Nishi Family › Compare › Autonomous Bug Resolver and Code Developer

Nishi Compare · capabilities, evidence and delivery

Autonomous Bug Resolver and Code Developer

Compare capabilities and delivery evidence. Source discovery, qualification, deployment and accepted outcomes are separate; an open contract names work still to verify.

The loop that turns the estate's own backlog and the field's benchmarks into verified, shipped fixes -- intake, localise, propose (local no-float tier, ox-alpha through the egress warden, Claude for judged work), reproduce, verify by executable oracle, ship, ledger, publish under a rigor envelope -- measured 2026-08-27 against Claude Code, Codex CLI, OpenHands and Live-SWE-agent and against the August-2026 research record

Overview

The purpose, declared position and evidence coverage of this domain. Source presence and completed acceptance are different measures.

Where we are. MEASURED 2026-08-27 against the catalogue: the self-localising fix loop nx_autofix_auto is BUILT and UNPROMOTED (325,366 B staged, nothing served); its intake gate nx_autofix_intake_gate and its SWE-bench-contract grader nx_swebench_local_gate are PROMOTED and UNREGISTERED; the evolution harness nx_evolve is PROMOTED and UNREGISTERED; the patch-cascade guard, the gated editor, the heal plane and the deep-research runner are registered and dark; the Verified grader nx_swebv_grade is registered and unauthorised. The published scoreboard at /compare/autograde reports 92 percent of our foreign-bug instances resolved under the SWE-bench contract with no denominator on the page, and the July record shows the loop measured 6 of 14 then 7 of 14 with all reverts exact. What IS live and proven: the byte-exact judge and bounded repair loop (coding CB0 and CB1), the mutation harness nx_gate_bite, the promote ruler nx_contentdiff, the ship loop nx_organ_ship with a journal, the failure-record ruler nx_failclass over 880 verdict streams, the egress warden nx_extllm and the external call nx_extllm_call with an egress ledger, and the local no-float Coder tiers. The estate's own boards are the task supply: 4,223 debt rows, 633 roster gates of which 73 read RED or absent or timed out, nine gates that have never gone red, 398 unwired functions, 3,476 inline magic numbers and 271 binaries behind their build -- every one with a machine oracle attached. The counts below are measured at emit.

Where we need to go. OPERATOR ASK (2026-08-27): an autonomous bug resolver and code developer built from what we have and what the August-2026 record shows, using ox-alpha, Claude and the local tools, with /compare as the emitter -- the second project after /compare/gen -- carrying proven evidence on BOTH boards as a mandatory condition, to a standard that would pass a real-world review and doctoral rigor. Deliverable: the loop as organs on the proven lanes, the estate's boards as intake, executable oracles as the only judge, a rigor envelope that refuses any bare number on either board, an external referee on decontaminated multilingual tasks, and a bounded self-improvement path whose proposals never auto-admit.

The unit. 1 u = one measured session-leg (organ plus gate plus live flip), the standing calibration. Local evidence: the trigger-first composer for /compare/gen went from record-mining to deployed in one leg on 2026-08-27; nx_extllm and nx_extllm_call landed in one leg each on 2026-08-25.

Where we are: 20 present rows, 20 open rungs. A loop that exists in pieces, most of them uncallable, graded by a number with no denominator. AHEAD on the oracle: byte-exact judge, mutation harness, promote ruler, failure-record classifier. BEHIND on scope (single function), on scale (1.5B local maker), on external standing (no leaderboard result), and on rigor of publication (no n, no interval, no harness hash).

Cost to an honest loop: 8.5 u. M0: the rigor envelope, the harness manifest, the intake plane, the null controls, the episode ledger, untrusted-input admission and the sandbox root. Version 0.1 is a loop whose every number can be refuted.

Cost to a resolver on the estate backlog: 6.5 u more. M1: reproduction tooth, verify chain, retrieval localiser, sampler tiers and best-of-N. 15 u cumulative = version 1.0, measured by the debt board's own closure count and by recurrence.

Cost to a grade-1 external number: 8 u more. M2: multi-file patch sets, the SWE-rebench V2 and SWE-bench Pro runner, the Aider-Polyglot runner. 23 u cumulative = version 2.0; the first number is whatever it is, published with its interval.

Cost to bounded self-improvement: 9.5 u more. M3: harness evolution as proposals scored on held-out tasks, bounded failure memory, chained maintenance, terms-gated distillation, prevention emission. 32.5 u cumulative = version 3.0.

Research bar. SWE-rebench V2, the post-cutoff multilingual task pool is measured on 32,079 executable real-world tasks across 20 languages from 3,617 repositories with pre-built images, unsound instances filtered by an ensemble of LLM judges [@swerebenchv2]. Theirs: the referee pool the field trains and evaluates on. Ours: AD14 runs a post-cutoff subset bench-only in an isolated host and publishes n, the interval and the harness hash; until then we have no number here and say so.

Research bar. SWE-bench Pro, the held-out long-horizon set is measured on 1,865 problems from 41 repositories in public, held-out and commercial splits; multi-file patches that take a professional hours to days; human-verified and contamination-resistant [@swebenchpro]. Theirs: Live-SWE-agent reports 45.8 percent on it without test-time scaling [liveswe]. Ours: no number; AD13 multi-file editing is the prerequisite and AD14 the run.

Research bar. ChainSWE, dependent bug chains in one codebase is measured on 304 chronologically chained issues across 54 projects; agents drop by up to 70 percent as chains lengthen [@chainswe]. Theirs: the axis nobody publishes a number on yet. Ours: AD18 publishes resolve rate per chain length over the debt board's own chains, where the same defect was filed five times in three weeks.

Research bar. Holistic Agent Leaderboard, the evaluation infrastructure is measured on 21,730 rollouts across 9 models and 9 benchmarks for about 40,000 dollars, with LLM-aided log inspection that found benchmark-gaming behaviours [@hal]. Theirs: standardised harness, three-dimensional analysis, all logs released. Ours: AD5 ships the episode ledger and AD2 the harness manifest so every run of ours is auditable the same way at zero dollars of inference on the sovereign tier.

Research bar. Darwin Godel Machine v2 is measured on empirically evaluated coding-agent evolution with a diverse archive. Theirs: pinned research baseline, not a present-day supremacy claim. Ours: Nishi outcome-loop comparison is not run; native research capture retained with SHA256235e599246ba5379eded0f138a7d00458b6d86bdab87e2bcd660bd22f8efa924.

Outcome-driven estate improvement: Retained requirement; autonomous superiority unproven. Operator direction2026-09-23 joins Compare intake, native APIs and workflows, independent experiments, inspected user outcomes, guarded release, measured recovery and ongoing improvement. Evaluate total token and resource cost, delivery speed, quality and accepted outcomes against matched baselines. Resource assurance remains first, with photoreal Beach scenes the product pilot. Intake1790132172; full requirements buildroot/knowledge/gates/estate-continuous-improvement-direction-20260923.md.

Latest recorded release

2026-09-03 · AD5

First work queue landed as DATA at knowledge/autodev_workqueue.conf, 7664 B sha256 14c8d96ea4c7b2fa. Three admitted rows, each carrying the machine oracle that decides it and the measurement that put it there; four exclusion rows, each carrying its refusal state and evidence. A SEED, NOT A CENSUS, and it says so.

Release history and work log

Release entries describe recorded changes; they do not establish that every capability passed evaluation.

27 of 41 capability source symbols found|13 of them authored stronger-claims|14 open|coverage 658/1000|adoption 13 full / 14 partial

Evidence profile — what the gaps on this board actually are

Measured by nx_swcompare_evidence, read back by nx_evprofile_lib. Every figure is a count with its denominator — there is deliberately no score, no grade and no percentage anywhere in this band, because a stored scalar is a field a seat can edit and a counted partition is not.

proven able to fail counts the gates that have a RECORDED RED — nx_gate_bite mutated the gate subject, rebuilt it, watched the gate go red, and that record is inside the shared TTL. never bitten is its complement over the same denominator: those gates ran and were green, and nothing has ever shown them able to detect anything, so their green is a statement about this run and not about the gate. green at 0/0 is a separate and much weaker observation — the gate printed GREEN on a zero denominator, so its own tooth counter says it examined nothing. A gate can be green, non-zero, and still never bitten; that is the common case and it is now visible instead of implied.

partition: grounded + unsupported = 27 vs present 27 · named + unnamed + withheld = 14 vs open 14 · both reconcile

liar-kill conj=GPqN · first failing conjunct: quorum

graded document: BUILDROOT tree, 23831 bytes · gates map: PRIMARY · stamped 2d 7h ago · source ../knowledge/status/evstamp_autodev.verdict

Gap class	What it is, and the work it names
GATE-FAILING	A declared gate ran and was not GREEN. The remedy is to FIX the gate it names. This is deliberately NOT merged with CLAIM-ONLY: one needs a gate written and the other needs a gate repaired, and a single merged number sends a seat at the wrong work.

This band reports the referee counts. The per-axis worklist rows — which axis is unsupported, which watch is ready to flip, which gap is unnamed — are printed by nx_swcompare_evidence autodev itself and are not carried on the stamp, so this page names the classes and the producer names the rows. That split is stated rather than hidden: a count without a worklist is not actionable, and this band is honest about which half of that it is.

Production map

Follow the dependencies, declared acceptance criteria and recorded priorities. Inspect source binding before treating a rank as executable work.

Ranking source binding: STALE_OR_INVALID_BINDING. Recorded priorities require current acceptance evidence and resource checks before execution.

Recorded ranking withheld

Retained ranking is not bound to the current plan and matrix: STALE_OR_INVALID_BINDING. Historical data remains retained. Refresh the ranking after resolving the input differences. Latest execution outcome is unverified.

Declared roadmap — contract, acceptance, executor, effort

Rung	Closes with	Definition of done (pre-declared)	Executor	Est.
Fix loop v0, self-localising, revert proven exact (AL0)	episode	LANDED 2026-07-16 as source and binary; BUILT-UNPROMOTED, which debt row autodev-loop-unpromoted below names	Organ	0 u
SWE-bench contract as the judge (AL1)	swb_resolved	LANDED, gate 4 of 4; PROMOTED-UNREGISTERED	Gate	0 u
Mutation harness, subject rebuilt per mutant (AL2)	gb_try	LANDED and LIVE; the oracle-of-the-oracle every rung below leans on	Organ	0 u
Egress warden and external call with a ledger row per call (AL3)	xl_call	LANDED and LIVE 2026-08-25 (extllm XL1 to XL3)	Organ	0 u
Rigor envelope in the gate base class (AD1)	gv_envelope	A published number is a struct -- value, n, a Wilson interval widened by hierarchical bootstrap over the task nesting, seed, harness hash, model pin, run date -- and the compare emit REFUSES a bare percentage for autodev AND gen [dontpassk] [statprecipice]. Done rule: the autograde 92-percent row and gen's G5 referee row both publish through it or both read REFUSED; neg-control: a hand-typed percent in api.json is refused by name; the interval width target is a conf row, never a literal	Gate	1.5 u
Harness disclosure manifest (AD2) after AD1	os_harness_manifest	The ship loop prints and journals a hash over tools, prompts, context builder, verifier, budgets and sampler tier for every result [harnessthesis]; done rule: two runs with different manifests are refused a direct comparison by the envelope; neg-control: a changed prompt file changes the hash	Organ	1 u
Intake plane from every board (AD3)	af_intake_board	debt, gate roster RED, failclass LATCHED, unwired, magic and artifactdrift BEHIND fold into one task plane, each task carrying its oracle and a RED-before receipt; the partition across sources sums to the plane count; chain order preserved [chainswe]; neg-control: a task whose oracle is GREEN before any fix is refused as unreproduced	Organ	1.5 u
Null controls (AD4) after AD1	af_null_control	Empty patch, revert patch, verbatim echo and replay of a prior solution run through the same judge on every batch and their scores print beside the resolver's [statprecipice]; done rule: every null control scores zero on the batch, and a batch where one does not is REFUSED as an oracle defect, never published	Gate	0.5 u
Episode ledger (AD5) after AD1	af_episode_ledger	One plane row per episode -- task, tier, samples, executions, verdict, tokens, wall, cost -- appended as decided so an interrupted run keeps what it finished; the api.json is emitted from the plane and from nothing else [hal]	Organ	1 u
Untrusted-input admission (AD6)	af_admit_untrusted	Deny by default: an external bug, patch or trajectory enters the tree or any training set only as DATA, provenance-pinned, and only if maintainer-merged upstream; eats debt rows 1784661149 and 1784661346; neg-control: a planted unmerged patch is refused by name; positive control: a pinned merged fix is admitted as data and never executed	Gate	1.5 u
Sandbox root (AD7)	af_sandbox_root	Every candidate builds and runs in an isolated scratch root under a time and resource limit and never touches the serving root [openhands] [dgm]; done rule: a candidate that writes outside its root is killed and the write is absent; neg-control: a fork bomb candidate is bounded	Organ	1 u
Reproduction tooth cogenerated with the fix (AD8) after AD3	af_repro_tooth	A fix ships with a gate row that fails on the pre-fix binary and passes on the post-fix one, and the planted mutant must die under nx_gate_bite [brtcogen]; done rule: fix rate with cogeneration is not lower than without, with non-overlapping intervals or a declared tie	Organ	1.5 u
Verify chain with an execution budget (AD9) after AD7	af_verify_chain	build, gate, bite, contentdiff, behaveprobe, magic and unwired ratchets as one pipeline; the execution budget per task class is a conf row derived from measured benefit [runornot]; done rule: a fix that adds a magic literal or an unwired function is REFUSED; neg-control: a fix that loses a live run is refused by the promote ruler	Organ	1.5 u
Retrieval-backed localiser (AD10) after AD3	af_localize_index	code index plus failclass plus gate stderr rank candidate functions; hit at 1 measured on a held-out split of the intake plane and published with its interval [adi]; done rule: hit at 1 exceeds the FNRES-only baseline with non-overlapping intervals or the leg stays UNWIRED	Organ	1.5 u
Sampler tiers as data (AD11) after AD6	af_sampler_tier	local 1.5B, ox-alpha through the warden, Claude for judged work, chosen per task class by a conf row; the free tier sees U bytes only [stealtheula] and the Claude tier's output never enters the corpus; done rule: the same task run on each tier lands in the same ledger with the tier named	Organ	1 u
Best-of-N against the free verifier (AD12) after AD11	fg_best_of_n	shared with coding CD2 and extllm XL5: pass at N monotone in N on the intake plane, NONE reported when nothing is GREEN [ttcagentic] [swezerohero]; AF_BON_N retires as a literal	Organ	1 u
Multi-file coordinated patch set (AD13) after AD9	fg_multi_file	shared with coding CD5: a task may name several organs, the judge builds the whole closure, a patch that breaks a third importer is refused [swebenchpro]	Organ	3 u
External referee run (AD14) after AD1,AD2,AD7	sg_rebench_run	a post-cutoff subset of SWE-rebench V2 and the SWE-bench Pro public split run bench-only in an isolated sandbox host under the untrusted-input rule, graded by our grader, published with n, interval and manifest hash [swerebenchv2] [swebenchpro] [swebenchcase]; done rule: the run publishes or it prints REFUSED with the missing conjunct named; a Verified-only run is a sanity check and never the headline	Gate	3 u
Aider-Polyglot runner (AD15) after AD1	fg_polyglot_bench	shared with coding CD6; publishes pass rate per language with N and the manifest hash	Organ	2 u
Harness evolution as proposals (AD16) after AD1,AD5,AD14	ev_harness_search	harness parameters are conf rows; the evolution harness proposes edits scored on a HELD-OUT split against a matched-budget test-time-scaling baseline [metaharness] [ahe] [harnessevalrethink] [lilharness]; a proposal lands on the pm intake plane and never auto-admits; done rule: a proposal is adopted only when it beats the baseline on held-out tasks with non-overlapping intervals; neg-control: a proposal that wins on the search split and loses on held-out is refused and the refusal is journaled	Organ	3 u
Bounded failure memory (AD17) after AD5	fc_failmem	a per-class failure library with an eviction bound as a conf row, read by the localiser and the prevention emitter [liveswe]; done rule: the store never exceeds its bound and eviction is journaled; neg-control: a store over bound refuses the write	Organ	1.5 u
Chained maintenance (AD18) after AD3,AD9	af_chain_run	dependent tasks on one organ family without a reset, resolve rate published per chain length with intervals [chainswe]; done rule: the drop with chain length is a published curve, not a hidden average	Organ	1.5 u
Terms-gated distillation (AD19) after AD5,AD6	xl_distill_corpus	shared with extllm XL9: verified episodes enter the sovereign corpus only from providers whose terms permit it, never from ox-alpha or Claude, and with every external sampler disabled the corpus must still strictly grow over a run [swezerohero]	Organ	1.5 u
Prevention emission (AD20) after AD17	fc_prevent_emit	every resolved class emits a new gate or lint; done rule: recurrence rate per class on the debt board falls after the emitter lands, measured over the whole board and never a sample; neg-control: a class with no emitted oracle shows no fall	Organ	2 u

Milestones

Milestone	Rungs	Cumulative
M0 · Honest loop	AD1,AD2,AD3,AD4,AD5,AD6,AD7	8.5 u
M1 · Resolver on the estate backlog	AD8,AD9,AD10,AD11,AD12	15 u
M2 · External referee	AD13,AD14,AD15	23 u
M3 · Bounded self-improvement	AD16,AD17,AD18,AD19,AD20	32.5 u

Rung	Closes with	Definition of done (pre-declared)	Executor	Est.
Design-choice migration: every adopted design row drives a whole-tree refactoring under the promote rulers (AD21) after AD9,AD18	af_design_migrate	OPERATOR 2026-09-22: as the language improves from the first bit up to the last, autodev keeps the estate aligned. For every design row on a board whose status is adopted and whose shape is mechanical (a shared helper replacing clones, a literal replaced by a derivation, a gate shape migrated, a syntax form rewritten), a migration recipe composes the incumbents (nx_oo_extract for consolidation with its byte-identical or behaveprobe proof, nx_magic derive for literals, nx_gate_migrate for gate shapes, nx_srcfmt for surface forms) over the WHOLE tree, never a sample; each run appends a before and after offender census per design id (files touched, files already converged and left byte-identical, files refused with the rule named) and writes the after count into that design row's measured-effect field on its board; the gate proves a planted offender is migrated, an already-converged file is untouched byte-for-byte, and a refused migration leaves the tree unchanged; the census is the ratchet: an adopted design row whose offender count rises is RED on the autodev board. First worklist: lang LD8 (six RIFF writers and the print, length, parse and concatenation clones), LD6 (value-named constants in the G2P and formant organs), LD5 (verdict-last across the gate fleet)	Organ	3 u

Ladder verdict. NOT DECLARED. This board names no dated best-in-class or frontier target and no rung roles (sotatarget and rungrole rows on its plan); the ranker labels it NO-LADDER until it does, and until then its rungs climb toward a target nobody has written down. Bars read NO-BAR against the current month 2026-10.

Bar freshness, counted
Class	Count	Note
bars	0	sotabar rows on the plan
fresh	0	the bar month is the current month
attested	0	an older bar with a this-month barscan reporting no newer entry
stale	0	a this-month scan saw a newer month the bar ignores
unattested	0	nobody looked this month
malformed	0	a sotabar row missing a field
duplicate	0	bar ids an earlier row already carries
bars, summed	0	RECONCILED: the parts sum to the total 0
current month	2026-10	the month every bar is judged against

Treatment verdict. NOT DECLARED. No rung on this board declares its three-step treatment (a treat row naming the MCP row, the API route, the primitive it composes, the workflow that composes it and the agent beat that runs it); the ranker labels every rung TREATMENT-UNDECLARED until it does.

Inspect a rung and its prerequisites

Declared nodes 25. Rank input binding: STALE_OR_INVALID_BINDING. Dependency order is authored. Implementation, acceptance evidence, authority and resource readiness are unverified. No action is recommended or dispatched here.

Plan SHA-256 ad3efb9b3a786d07ea12f3899fd2bbbcf1dff4272b175853c8082cb8f6a84480. Target rows 0; role rows 0. Existing risks and release worklog retain their own scope; no node completion is inferred.

Runtime observation coverage: 0 of 25 rungs have current, identity-bound execution observations. Observation may report RED. Accepted completion remains unknown; source presence, qualification and deployment do not establish it.

Use Enter or Space on a rung to inspect its contract. Prerequisite links locate another rung in this list; open its summary to inspect it. Estimates are authored effort, not forecasts.

AL0 — Fix loop v0, self-localising, revert proven exact
Prerequisites: None declared; this does not establish execution eligibility.
Contract: episode
Acceptance: LANDED 2026-07-16 as source and binary; BUILT-UNPROMOTED, which debt row autodev-loop-unpromoted below names
Authored effort: 0. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AL1 — SWE-bench contract as the judge
Prerequisites: None declared; this does not establish execution eligibility.
Contract: swb_resolved
Acceptance: LANDED, gate 4 of 4; PROMOTED-UNREGISTERED
Authored effort: 0. Executor kind: Gate. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AL2 — Mutation harness, subject rebuilt per mutant
Prerequisites: None declared; this does not establish execution eligibility.
Contract: gb_try
Acceptance: LANDED and LIVE; the oracle-of-the-oracle every rung below leans on
Authored effort: 0. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AL3 — Egress warden and external call with a ledger row per call
Prerequisites: None declared; this does not establish execution eligibility.
Contract: xl_call
Acceptance: LANDED and LIVE 2026-08-25 (extllm XL1 to XL3)
Authored effort: 0. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD1 — Rigor envelope in the gate base class
Prerequisites: None declared; this does not establish execution eligibility.
Contract: gv_envelope
Acceptance: A published number is a struct -- value, n, a Wilson interval widened by hierarchical bootstrap over the task nesting, seed, harness hash, model pin, run date -- and the compare emit REFUSES a bare percentage for autodev AND gen [@dontpassk] [@statprecipice]. Done rule: the autograde 92-percent row and gen's G5 referee row both publish through it or both read REFUSED; neg-control: a hand-typed percent in api.json is refused by name; the interval width target is a conf row, never a literal
Authored effort: 1.5. Executor kind: Gate. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD2 — Harness disclosure manifest
Prerequisites: AD1 (acceptance unverified)
Contract: os_harness_manifest
Acceptance: The ship loop prints and journals a hash over tools, prompts, context builder, verifier, budgets and sampler tier for every result [@harnessthesis]; done rule: two runs with different manifests are refused a direct comparison by the envelope; neg-control: a changed prompt file changes the hash
Authored effort: 1. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD3 — Intake plane from every board
Prerequisites: None declared; this does not establish execution eligibility.
Contract: af_intake_board
Acceptance: debt, gate roster RED, failclass LATCHED, unwired, magic and artifactdrift BEHIND fold into one task plane, each task carrying its oracle and a RED-before receipt; the partition across sources sums to the plane count; chain order preserved [@chainswe]; neg-control: a task whose oracle is GREEN before any fix is refused as unreproduced
Authored effort: 1.5. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD4 — Null controls
Prerequisites: AD1 (acceptance unverified)
Contract: af_null_control
Acceptance: Empty patch, revert patch, verbatim echo and replay of a prior solution run through the same judge on every batch and their scores print beside the resolver's [@statprecipice]; done rule: every null control scores zero on the batch, and a batch where one does not is REFUSED as an oracle defect, never published
Authored effort: 0.5. Executor kind: Gate. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD5 — Episode ledger
Prerequisites: AD1 (acceptance unverified)
Contract: af_episode_ledger
Acceptance: One plane row per episode -- task, tier, samples, executions, verdict, tokens, wall, cost -- appended as decided so an interrupted run keeps what it finished; the api.json is emitted from the plane and from nothing else [@hal]
Authored effort: 1. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD6 — Untrusted-input admission
Prerequisites: None declared; this does not establish execution eligibility.
Contract: af_admit_untrusted
Acceptance: Deny by default: an external bug, patch or trajectory enters the tree or any training set only as DATA, provenance-pinned, and only if maintainer-merged upstream; eats debt rows 1784661149 and 1784661346; neg-control: a planted unmerged patch is refused by name; positive control: a pinned merged fix is admitted as data and never executed
Authored effort: 1.5. Executor kind: Gate. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD7 — Sandbox root
Prerequisites: None declared; this does not establish execution eligibility.
Contract: af_sandbox_root
Acceptance: Every candidate builds and runs in an isolated scratch root under a time and resource limit and never touches the serving root [@openhands] [@dgm]; done rule: a candidate that writes outside its root is killed and the write is absent; neg-control: a fork bomb candidate is bounded
Authored effort: 1. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD8 — Reproduction tooth cogenerated with the fix
Prerequisites: AD3 (acceptance unverified)
Contract: af_repro_tooth
Acceptance: A fix ships with a gate row that fails on the pre-fix binary and passes on the post-fix one, and the planted mutant must die under nx_gate_bite [@brtcogen]; done rule: fix rate with cogeneration is not lower than without, with non-overlapping intervals or a declared tie
Authored effort: 1.5. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD9 — Verify chain with an execution budget
Prerequisites: AD7 (acceptance unverified)
Contract: af_verify_chain
Acceptance: build, gate, bite, contentdiff, behaveprobe, magic and unwired ratchets as one pipeline; the execution budget per task class is a conf row derived from measured benefit [@runornot]; done rule: a fix that adds a magic literal or an unwired function is REFUSED; neg-control: a fix that loses a live run is refused by the promote ruler
Authored effort: 1.5. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD10 — Retrieval-backed localiser
Prerequisites: AD3 (acceptance unverified)
Contract: af_localize_index
Acceptance: code index plus failclass plus gate stderr rank candidate functions; hit at 1 measured on a held-out split of the intake plane and published with its interval [@adi]; done rule: hit at 1 exceeds the FNRES-only baseline with non-overlapping intervals or the leg stays UNWIRED
Authored effort: 1.5. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD11 — Sampler tiers as data
Prerequisites: AD6 (acceptance unverified)
Contract: af_sampler_tier
Acceptance: local 1.5B, ox-alpha through the warden, Claude for judged work, chosen per task class by a conf row; the free tier sees U bytes only [@stealtheula] and the Claude tier's output never enters the corpus; done rule: the same task run on each tier lands in the same ledger with the tier named
Authored effort: 1. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD12 — Best-of-N against the free verifier
Prerequisites: AD11 (acceptance unverified)
Contract: fg_best_of_n
Acceptance: shared with coding CD2 and extllm XL5: pass at N monotone in N on the intake plane, NONE reported when nothing is GREEN [@ttcagentic] [@swezerohero]; AF_BON_N retires as a literal
Authored effort: 1. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD13 — Multi-file coordinated patch set
Prerequisites: AD9 (acceptance unverified)
Contract: fg_multi_file
Acceptance: shared with coding CD5: a task may name several organs, the judge builds the whole closure, a patch that breaks a third importer is refused [@swebenchpro]
Authored effort: 3. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD14 — External referee run
Prerequisites: AD1 (acceptance unverified), AD2 (acceptance unverified), AD7 (acceptance unverified)
Contract: sg_rebench_run
Acceptance: a post-cutoff subset of SWE-rebench V2 and the SWE-bench Pro public split run bench-only in an isolated sandbox host under the untrusted-input rule, graded by our grader, published with n, interval and manifest hash [@swerebenchv2] [@swebenchpro] [@swebenchcase]; done rule: the run publishes or it prints REFUSED with the missing conjunct named; a Verified-only run is a sanity check and never the headline
Authored effort: 3. Executor kind: Gate. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD15 — Aider-Polyglot runner
Prerequisites: AD1 (acceptance unverified)
Contract: fg_polyglot_bench
Acceptance: shared with coding CD6; publishes pass rate per language with N and the manifest hash
Authored effort: 2. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD16 — Harness evolution as proposals
Prerequisites: AD1 (acceptance unverified), AD5 (acceptance unverified), AD14 (acceptance unverified)
Contract: ev_harness_search
Acceptance: harness parameters are conf rows; the evolution harness proposes edits scored on a HELD-OUT split against a matched-budget test-time-scaling baseline [@metaharness] [@ahe] [@harnessevalrethink] [@lilharness]; a proposal lands on the pm intake plane and never auto-admits; done rule: a proposal is adopted only when it beats the baseline on held-out tasks with non-overlapping intervals; neg-control: a proposal that wins on the search split and loses on held-out is refused and the refusal is journaled
Authored effort: 3. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD17 — Bounded failure memory
Prerequisites: AD5 (acceptance unverified)
Contract: fc_failmem
Acceptance: a per-class failure library with an eviction bound as a conf row, read by the localiser and the prevention emitter [@liveswe]; done rule: the store never exceeds its bound and eviction is journaled; neg-control: a store over bound refuses the write
Authored effort: 1.5. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD18 — Chained maintenance
Prerequisites: AD3 (acceptance unverified), AD9 (acceptance unverified)
Contract: af_chain_run
Acceptance: dependent tasks on one organ family without a reset, resolve rate published per chain length with intervals [@chainswe]; done rule: the drop with chain length is a published curve, not a hidden average
Authored effort: 1.5. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD19 — Terms-gated distillation
Prerequisites: AD5 (acceptance unverified), AD6 (acceptance unverified)
Contract: xl_distill_corpus
Acceptance: shared with extllm XL9: verified episodes enter the sovereign corpus only from providers whose terms permit it, never from ox-alpha or Claude, and with every external sampler disabled the corpus must still strictly grow over a run [@swezerohero]
Authored effort: 1.5. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD20 — Prevention emission
Prerequisites: AD17 (acceptance unverified)
Contract: fc_prevent_emit
Acceptance: every resolved class emits a new gate or lint; done rule: recurrence rate per class on the debt board falls after the emitter lands, measured over the whole board and never a sample; neg-control: a class with no emitted oracle shows no fall
Authored effort: 2. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
AD21 — Design-choice migration: every adopted design row drives a whole-tree refactoring under the promote rulers
Prerequisites: AD9 (acceptance unverified), AD18 (acceptance unverified)
Contract: af_design_migrate
Acceptance: OPERATOR 2026-09-22: as the language improves from the first bit up to the last, autodev keeps the estate aligned. For every design row on a board whose status is adopted and whose shape is mechanical (a shared helper replacing clones, a literal replaced by a derivation, a gate shape migrated, a syntax form rewritten), a migration recipe composes the incumbents (nx_oo_extract for consolidation with its byte-identical or behaveprobe proof, nx_magic derive for literals, nx_gate_migrate for gate shapes, nx_srcfmt for surface forms) over the WHOLE tree, never a sample; each run appends a before and after offender census per design id (files touched, files already converged and left byte-identical, files refused with the rule named) and writes the after count into that design row's measured-effect field on its board; the gate proves a planted offender is migrated, an already-converged file is untouched byte-for-byte, and a refused migration leaves the tree unchanged; the census is the ratchet: an adopted design row whose offender count rises is RED on the autodev board. First worklist: lang LD8 (six RIFF writers and the print, length, parse and concatenation clones), LD6 (value-named constants in the G2P and formant organs), LD5 (verdict-last across the gate fleet)
Authored effort: 3. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.

Learning and practice paths

No structured learning path is declared for this plan. Existing research, roadmap and worklog remain available above.

Rung dossiers — every record that bears on a rung, in one place

Joined by symbol, never typed. Each dossier is assembled on every publish from the rows this board already carries: the plan's rung row, the matrix row that carries its contract symbol (its status re-measured by the one symbol ruler), the gate the organ declares and the verdict the evidence referee recorded on its last run, the reference marks in the rung text, the design rows that name the rung and the work log filed against it. The capability matrix links each cell to its dossier and each dossier links back to its cell and forward to the rungs it stands on; a dependency naming no rung on this plan is written as UNRESOLVED rather than linked. Absences are named: NO-MATRIX-ROW, NO-GATE-DECLARED, UNRECORDED.

Gate verdicts: UNSTAMPED — no per-gate verdict artifact is published for this domain yet (nx_swcompare_evidence autodev writes it beside its stamp); every gate below reads UNRECORDED until it runs.

Gate declarations: knowledge/organ_gate.conf read from this working directory (1 rows) and from the estate root (196 rows); rows in both counted once: 0; declarations read: 197. Every declaration in either copy counts, so a stub beside the publisher can never hide the root.

AL0 Fix loop v0, self-localising, revert proven exact · cell PRESENT · gate NO-GATE-DECLARED

Contract	`episode` · executor Organ · 0 u · stands on nothing
Definition of done	LANDED 2026-07-16 as source and binary; BUILT-UNPROMOTED, which debt row autodev-loop-unpromoted below names
Matrix cell	PROPOSE: the fix-loop episode -- localise, generate, apply, fresh compile plus run as the sole judge, revert proven exact · `runtime/nx_autofix_auto.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AL1 SWE-bench contract as the judge · cell PRESENT · gate NO-GATE-DECLARED

Contract	`swb_resolved` · executor Gate · 0 u · stands on nothing
Definition of done	LANDED, gate 4 of 4; PROMOTED-UNREGISTERED
Matrix cell	VERIFY: the SWE-bench contract as the sole judge -- FAIL_TO_PASS flips AND PASS_TO_PASS holds · `runtime/nx_swebench_local_gate.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_swebench_local_gate`; declare `nx_swebench_local_gate TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AL2 Mutation harness, subject rebuilt per mutant · cell PRESENT · gate NO-GATE-DECLARED

Contract	`gb_try` · executor Organ · 0 u · stands on nothing
Definition of done	LANDED and LIVE; the oracle-of-the-oracle every rung below leans on
Matrix cell	VERIFY: non-vacuity by mutation -- can this gate ever fail? the subject is rebuilt and deployed per mutant and restored by hash · `runtime/_hdl_build/nx_gate_bite.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_gate_bite`; declare `nx_gate_bite TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AL3 Egress warden and external call with a ledger row per call · cell PRESENT · gate NO-GATE-DECLARED

Contract	`xl_call` · executor Organ · 0 u · stands on nothing
Definition of done	LANDED and LIVE 2026-08-25 (extllm XL1 to XL3)
Matrix cell	SAMPLER: an external frontier sampler behind the warden -- ox-alpha, 1,048,576-token context, free · `runtime/nx_extllm_call.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_extllm_call`; declare `nx_extllm_call TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD1 Rigor envelope in the gate base class · cell PRESENT · gate NO-GATE-DECLARED

Contract	`gv_envelope` · executor Gate · 1.5 u · stands on nothing
Definition of done	A published number is a struct -- value, n, a Wilson interval widened by hierarchical bootstrap over the task nesting, seed, harness hash, model pin, run date -- and the compare emit REFUSES a bare percentage for autodev AND gen [dontpassk] [statprecipice]. Done rule: the autograde 92-percent row and gen's G5 referee row both publish through it or both read REFUSED; neg-control: a hand-typed percent in api.json is refused by name; the interval width target is a conf row, never a literal
Matrix cell	RIGOR: every published number carries n, a Wilson interval with hierarchical bootstrap over the nested task structure, seed, harness hash and model pin -- and the emit REFUSES a bare percentage · `runtime/nx_gate_verdict.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_gate_verdict`; declare `nx_gate_verdict TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	2026-09-03 land gv_bare_rate WIRED into gv_verdict, the emitter every gate calls. buildroot/runtime/nx_gate_verdict.nx 35612 -> 38635 B, sha256 799e453ad2781a8b -> 6c6a488fac255b0a. New func gv_note_bare_rate scans the note BEFORE any verdict byte is emitted, so the announcement can never displace the verdict line from the tail where gv_last_line anchors. SILENT WHEN CLEAN by construction, so stdout stays byte-identical for every gate without the defect and no judge-equivalence proof is invalidated. It ANNOUNCES rather than refuses, chosen not conceded: gv_verdict's return value IS the fleet verdict, so a refusing base class would turn every gate whose note quotes a rate RED in a single edit. 2026-09-03 measure Wiring proven on the ARTIFACT, never the receipt. /api/build BUILT 57840 with src_path naming nx_rigor_envelope_gate and src_stable=1; nx_contentdiff lost_from_live=0 lost_permil=0 symmetry=ONE-WAY-GAIN verdict=GREEN 3 of 3, its GAINED list naming the new string NX-RIGOR bare-rate-in-note byte_offset= so the code is provably IN the binary; nx_behaveprobe first_diff_off=-1 verdict=IDENTICAL evidence=STRONG; PROMOTED sha256 b292fcf0be57845b prev=1; /api/gate_run against the LIVE promoted binary exit_code=0 verdict=GREEN capture_truncated=0. 2026-09-03 measure Adoption census taken BEFORE the edit, coverage_complete=1 corpus_complete=1 over files=23592: gv_bare_rate matches=8, every one its own definition, its own comment, or nx_rigor_envelope_gate which is its own gate; gv_envelope_check matches=1, its definition alone. The refusing half of AD1 was built, gate-proven, and unreachable from every production emitter in the estate. 2026-09-03 lesson A RULER PLACED IN THE BASE CLASS IS NOT THEREBY ADOPTED. nx_gate_verdict's own rigor header argues that a ruler in a sibling lib is adopted at advice rates while a ruler in the base class is adopted by every gate that imports it -- and gv_bare_rate then sat in that very base class with ZERO callers. The adoption argument was made and the wiring was not done. 2026-09-03 retract Two recommendations withdrawn, the second by measurement taken this session. The first, build nx_autodev_qa because no adjudicator exists, was already withdrawn as a top-8 ranked search read as a population. Its replacement is ALSO refuted: nx_envelope_audit is NOT the adjudicator, not even partly. Reading its source shows a SOURCE-TEXT scanner over the registered tool_allowlist classifying rows DECLARING vs SILENT-CAP-CAND vs UNBOUNDED-OR-NA vs NAME-PROBE-MISS for the SCALE envelope, meaning bounded reads, _CAP and _MAX consts and silent truncation, and its own header states that a detector reading source text cannot enforce a contract defined on runtime output. The rank-2 hit was a HOMONYM: envelope means bounded-read-declaration in nx_envelope_audit and rate-plus-denominator-plus-interval in gv_envelope_check. Two organs, one word, unrelated subjects.

AD2 Harness disclosure manifest · cell PRESENT · gate UNRECORDED

Contract	`os_harness_manifest` · executor Organ · 1 u · stands on AD1
Definition of done	The ship loop prints and journals a hash over tools, prompts, context builder, verifier, budgets and sampler tier for every result [harnessthesis]; done rule: two runs with different manifests are refused a direct comparison by the envelope; neg-control: a changed prompt file changes the hash
Matrix cell	RIGOR: harness disclosure manifest -- tools, prompts, context builder, verifier, budgets and sampler tier, hashed and printed beside every result · `runtime/nx_organ_ship.nx` · PRESENT decl
Gate	`nx_organ_ship_gate` · UNRECORDED — no recorded verdict for this gate (it is declared for the organ but the evidence referee's last run did not record it: it is not on this board's gates map, or the referee has not run since the per-gate record shipped) `nx_organ_ship_behavior_gate` · UNRECORDED — no recorded verdict for this gate (it is declared for the organ but the evidence referee's last run did not record it: it is not on this board's gates map, or the referee has not run since the per-gate record shipped)
Work log	no log row names this rung yet

AD3 Intake plane from every board · cell PRESENT · gate NO-GATE-DECLARED

Contract	`af_intake_board` · executor Organ · 1.5 u · stands on nothing
Definition of done	debt, gate roster RED, failclass LATCHED, unwired, magic and artifactdrift BEHIND fold into one task plane, each task carrying its oracle and a RED-before receipt; the partition across sources sums to the plane count; chain order preserved [chainswe]; neg-control: a task whose oracle is GREEN before any fix is refused as unreproduced
Matrix cell	INTAKE: one task plane from every board -- debt, gate roster RED, failclass LATCHED, unwired, magic, artifactdrift BEHIND -- each task carrying its executable oracle and a RED-before receipt, partition summing · `runtime/nx_autofix_auto.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD4 Null controls · cell PRESENT · gate NO-GATE-DECLARED

Contract	`af_null_control` · executor Gate · 0.5 u · stands on AD1
Definition of done	Empty patch, revert patch, verbatim echo and replay of a prior solution run through the same judge on every batch and their scores print beside the resolver's [statprecipice]; done rule: every null control scores zero on the batch, and a batch where one does not is REFUSED as an oracle defect, never published
Matrix cell	CONTROL: null resolvers -- empty patch, revert patch, verbatim echo, replay of a prior solution -- must score zero, printed beside every run · `runtime/nx_autofix_auto.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD5 Episode ledger · cell PRESENT · gate NO-GATE-DECLARED

Contract	`af_episode_ledger` · executor Organ · 1 u · stands on AD1
Definition of done	One plane row per episode -- task, tier, samples, executions, verdict, tokens, wall, cost -- appended as decided so an interrupted run keeps what it finished; the api.json is emitted from the plane and from nothing else [hal]
Matrix cell	LEDGER: one row per episode -- task, tier, samples, executions, verdict, tokens, wall, cost -- and the page reads the plane, never a hand number · `runtime/nx_autofix_auto.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	2026-09-03 land QA admission contract landed as DATA at knowledge/autodev_admit.conf, 7909 B sha256 6c861099c41ad4cd. Nine NAMED states -- ADMIT, four REFUSE and four UNPROVEN -- each naming its DECIDER organ and the PASSTHROUGH field whose bytes are copied verbatim into the record, because a hand-copied paraphrase of another organ's verdict is a duplicate ruler made of prose. Every UNPROVEN state ABSTAINS and can never grant an ADMIT. UNCONSUMED until an organ names it, and the file says so rather than claiming enforcement it does not have. 2026-09-03 land First work queue landed as DATA at knowledge/autodev_workqueue.conf, 7664 B sha256 14c8d96ea4c7b2fa. Three admitted rows, each carrying the machine oracle that decides it and the measurement that put it there; four exclusion rows, each carrying its refusal state and evidence. A SEED, NOT A CENSUS, and it says so. 2026-09-03 lesson nx_organ_ship prints prove=NO-GATE-FOUND UNPROVEN -- a NAMED absence, NOT a pass -- at nx_organ_ship.nx:564 and :565 and carries it to the final line at :794, AND PROMOTES ANYWAY. Correct for a human ship and catastrophic for an autonomous worker, because the difference is not the message but who is left to act on it. Three organs exist because this already happened: nx_raidwatch_gate, nx_search_census_gate, and nx_seat_gate, the last recording that a 29-tooth gate sat ONE ARGUMENT AWAY while the ship reported nothing proven. 2026-09-03 queue Next rung is the adjudicator that READS knowledge/autodev_admit.conf, plus the census that turns the three-row seed queue into a population. Neither is claimed here.

AD6 Untrusted-input admission · cell PRESENT · gate NO-GATE-DECLARED

Contract	`af_admit_untrusted` · executor Gate · 1.5 u · stands on nothing
Definition of done	Deny by default: an external bug, patch or trajectory enters the tree or any training set only as DATA, provenance-pinned, and only if maintainer-merged upstream; eats debt rows 1784661149 and 1784661346; neg-control: a planted unmerged patch is refused by name; positive control: a pinned merged fix is admitted as data and never executed
Matrix cell	GUARD: untrusted-input admission -- no external bug, patch or trajectory enters the tree or a training set unless it is DATA-ONLY, provenance-pinned and maintainer-merged; deny by default · `runtime/nx_autofix_auto.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	2026-09-02 measure 2026-09-02: nx_autofix_intake_gate read passed 3/5 verdict=RED on the NAS (15 proposed, 15 NO FNRES ROWS -> DISCARDED, kept=0) and the domain sat on the CLAIMED-BUT-RED list that blocks every deploy. Root cause is not the intake: every sandbox compile goes through _offc/nx_sov_build_run.elf, which exits SBR_ADMIT_REFUSED=6 under the I/O storm, so the grader capture was empty because nothing was BUILT, and the gate read an unbuilt instance as a non-reproducing one. Same class as the sclass dispatch gate and the referee deadline fixed earlier today 2026-09-02 queue Source fix landed (10607 B): AIG_BUILDER_ADMIT_REFUSED=6 sorts a refused compile into an unexamined bucket (never discarded, printed BUILDER-ADMISSION-REFUSED), reached = kept+discarded >= 1 is a gv_need precondition, T2/T3/T5 are asserted only when reached, the partition line carries unexamined= and still sums to proposed. Expected on a saturated box: SKIP exit 3 with T1 and T4 still run; expected on a calm box: unchanged 5/5. Build queued behind admission (governor QUEUE at procs_running=8/8; /api/build enqueued); promote, gate_run and the autodev referee re-run owed

AD7 Sandbox root · cell PRESENT · gate NO-GATE-DECLARED

Contract	`af_sandbox_root` · executor Organ · 1 u · stands on nothing
Definition of done	Every candidate builds and runs in an isolated scratch root under a time and resource limit and never touches the serving root [openhands] [dgm]; done rule: a candidate that writes outside its root is killed and the write is absent; neg-control: a fork bomb candidate is bounded
Matrix cell	SANDBOX: every candidate builds and runs in an isolated scratch root with time and resource limits, never the serving root · `runtime/nx_autofix_auto.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD8 Reproduction tooth cogenerated with the fix · cell WATCHING · gate NO-GATE-DECLARED

Contract	`af_repro_tooth` · executor Organ · 1.5 u · stands on AD3
Definition of done	A fix ships with a gate row that fails on the pre-fix binary and passes on the post-fix one, and the planted mutant must die under nx_gate_bite [brtcogen]; done rule: fix rate with cogeneration is not lower than without, with non-overlapping intervals or a declared tie
Matrix cell	REPRO: a bug-reproduction tooth cogenerated with every fix -- a neg-control that fails before and passes after, and the planted mutant must die · `runtime/nx_autofix_auto.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD9 Verify chain with an execution budget · cell WATCHING · gate NO-GATE-DECLARED

Contract	`af_verify_chain` · executor Organ · 1.5 u · stands on AD7
Definition of done	build, gate, bite, contentdiff, behaveprobe, magic and unwired ratchets as one pipeline; the execution budget per task class is a conf row derived from measured benefit [runornot]; done rule: a fix that adds a magic literal or an unwired function is REFUSED; neg-control: a fix that loses a live run is refused by the promote ruler
Matrix cell	VERIFY: the verify chain as one pipeline -- build, gate, bite, contentdiff, behaveprobe, magic and unwired ratchets -- with an execution budget measured per task class · `runtime/nx_autofix_auto.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	2026-09-12 operator operator-unstall-20260912: Resume Nishi autodevelopment through existing nx_autofix_auto and nx_autofix_lib with a reusable MCP/API workflow, alongside Beach, Browser, Video and Search. Fresh intake job 1789237042 succeeds: total=694 gates_red=155 drift=38 magic=501 debt_refused_no_oracle=4450 partition=RECONCILES. CURRENT BLOCKER job 1789237101: all 15 existing proposal sandbox runs exit 7 REFUSED-TREE-CANON INPUT, compiler not launched, yet intake counts discarded=15/unexamined=0 and returns 4/6 RED. This supersedes the old load-exit-6 diagnosis. Smallest repair: carry validated canonical build context into the existing sandbox via the build owner's supported route and classify unexecuted/refused runs as UNEXAMINED with exact reason. Acceptance: same 15-proposal intake builds in isolation, reproducing bugs retained, equivalent mutant discarded, missing canon abstains, tree unchanged, named gate and no-provider-call receipt; then exercise existing admission contract before any autonomous promotion. Registration and intake invocation do not establish an adopted repair loop.

AD10 Retrieval-backed localiser · cell WATCHING · gate NO-GATE-DECLARED

Contract	`af_localize_index` · executor Organ · 1.5 u · stands on AD3
Definition of done	code index plus failclass plus gate stderr rank candidate functions; hit at 1 measured on a held-out split of the intake plane and published with its interval [adi]; done rule: hit at 1 exceeds the FNRES-only baseline with non-overlapping intervals or the leg stays UNWIRED
Matrix cell	LOCALIZE: retrieval-backed localisation -- code index plus failclass plus gate stderr -- with hit at 1 measured on held-out tasks · `runtime/nx_autofix_auto.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD11 Sampler tiers as data · cell WATCHING · gate NO-GATE-DECLARED

Contract	`af_sampler_tier` · executor Organ · 1 u · stands on AD6
Definition of done	local 1.5B, ox-alpha through the warden, Claude for judged work, chosen per task class by a conf row; the free tier sees U bytes only [stealtheula] and the Claude tier's output never enters the corpus; done rule: the same task run on each tier lands in the same ledger with the tier named
Matrix cell	SAMPLER: sampler tiers as DATA -- local 1.5B, ox-alpha through the warden, Claude for judged work -- selected per task class by a conf row, never by code · `runtime/nx_autofix_auto.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD12 Best-of-N against the free verifier · cell WATCHING · gate NO-GATE-DECLARED

Contract	`fg_best_of_n` · executor Organ · 1 u · stands on AD11
Definition of done	shared with coding CD2 and extllm XL5: pass at N monotone in N on the intake plane, NONE reported when nothing is GREEN [ttcagentic] [swezerohero]; AF_BON_N retires as a literal
Matrix cell	PROPOSE: best-of-N against the free verifier · `runtime/nx_forge.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_forge`; declare `nx_forge TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD13 Multi-file coordinated patch set · cell WATCHING · gate NO-GATE-DECLARED

Contract	`fg_multi_file` · executor Organ · 3 u · stands on AD9
Definition of done	shared with coding CD5: a task may name several organs, the judge builds the whole closure, a patch that breaks a third importer is refused [swebenchpro]
Matrix cell	PROPOSE: multi-file coordinated patch set, the judge builds the whole closure · `runtime/nx_forge.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_forge`; declare `nx_forge TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD14 External referee run · cell WATCHING · gate NO-GATE-DECLARED

Contract	`sg_rebench_run` · executor Gate · 3 u · stands on AD1, AD2, AD7
Definition of done	a post-cutoff subset of SWE-rebench V2 and the SWE-bench Pro public split run bench-only in an isolated sandbox host under the untrusted-input rule, graded by our grader, published with n, interval and manifest hash [swerebenchv2] [swebenchpro] [swebenchcase]; done rule: the run publishes or it prints REFUSED with the missing conjunct named; a Verified-only run is a sanity check and never the headline
Matrix cell	REFEREE: SWE-rebench V2 and SWE-bench Pro post-cutoff multilingual subset, run bench-only in an isolated sandbox host, graded by our grader · `runtime/_hdl_build/nx_swebv_grade.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_swebv_grade`; declare `nx_swebv_grade TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD15 Aider-Polyglot runner · cell WATCHING · gate NO-GATE-DECLARED

Contract	`fg_polyglot_bench` · executor Organ · 2 u · stands on AD1
Definition of done	shared with coding CD6; publishes pass rate per language with N and the manifest hash
Matrix cell	REFEREE: the Aider-Polyglot runner · `runtime/nx_forge.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_forge`; declare `nx_forge TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD16 Harness evolution as proposals · cell WATCHING · gate NO-GATE-DECLARED

Contract	`ev_harness_search` · executor Organ · 3 u · stands on AD1, AD5, AD14
Definition of done	harness parameters are conf rows; the evolution harness proposes edits scored on a HELD-OUT split against a matched-budget test-time-scaling baseline [metaharness] [ahe] [harnessevalrethink] [lilharness]; a proposal lands on the pm intake plane and never auto-admits; done rule: a proposal is adopted only when it beats the baseline on held-out tasks with non-overlapping intervals; neg-control: a proposal that wins on the search split and loses on held-out is refused and the refusal is journaled
Matrix cell	EVOLVE: harness evolution as PROPOSALS -- harness parameters are conf rows, the loop proposes edits scored on a HELD-OUT split against a matched-budget test-time-scaling baseline, and nothing auto-admits · `runtime/_hdl_build/nx_evolve.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_evolve`; declare `nx_evolve TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	2026-09-23 operator Estate-wide continuous improvement must deliver both an inspectable human outcome and an accountable machine feedback loop. Native research captures retained and hashes checked for DGMv2 and the independent research case-study evaluation2607.27191. Candidate and evaluator changes require separate review; recovery must be exercised. No achieved superiority, autonomous scientist or positive swarmROI is claimed. Existing loadgov LV21/LV25, PM intake1790132172 and shared publication owners are the initial integration path, not a duplicate service. 2026-09-23 queue INTAKE 1790132172 routed by nx_intake_route (nishi-review-via-codex, epoch 1790146428) from operator-via-codex: Estate-wide continuous improvement: pair an inspectable human outcome with an accountable recurring machine loop. Compare intake through native experiments, independent evaluation, guarded release, re -- sponsor intake-1790132172 plus1

AD17 Bounded failure memory · cell WATCHING · gate UNRECORDED

Contract	`fc_failmem` · executor Organ · 1.5 u · stands on AD5
Definition of done	a per-class failure library with an eviction bound as a conf row, read by the localiser and the prevention emitter [liveswe]; done rule: the store never exceeds its bound and eviction is journaled; neg-control: a store over bound refuses the write
Matrix cell	MEMORY: bounded failure memory -- a per-class failure library with eviction, feeding prevention rather than prompts · `runtime/nx_failclass.nx` · WATCHING decl
Gate	`nx_refusal_shape_gate` · UNRECORDED — no recorded verdict for this gate (it is declared for the organ but the evidence referee's last run did not record it: it is not on this board's gates map, or the referee has not run since the per-gate record shipped)
Work log	no log row names this rung yet

AD18 Chained maintenance · cell WATCHING · gate NO-GATE-DECLARED

Contract	`af_chain_run` · executor Organ · 1.5 u · stands on AD3, AD9
Definition of done	dependent tasks on one organ family without a reset, resolve rate published per chain length with intervals [chainswe]; done rule: the drop with chain length is a published curve, not a hidden average
Matrix cell	CHAIN: sequential dependent tasks on one organ family without a reset, resolve rate published per chain length · `runtime/nx_autofix_auto.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_autofix_auto`; declare `nx_autofix_auto TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD19 Terms-gated distillation · cell WATCHING · gate NO-GATE-DECLARED

Contract	`xl_distill_corpus` · executor Organ · 1.5 u · stands on AD5, AD6
Definition of done	shared with extllm XL9: verified episodes enter the sovereign corpus only from providers whose terms permit it, never from ox-alpha or Claude, and with every external sampler disabled the corpus must still strictly grow over a run [swezerohero]
Matrix cell	DISTILL: terms-gated harvest of verified episodes into the sovereign corpus -- never from ox-alpha or Claude · `runtime/nx_extllm.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_extllm`; declare `nx_extllm TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

AD20 Prevention emission · cell WATCHING · gate UNRECORDED

Contract	`fc_prevent_emit` · executor Organ · 2 u · stands on AD17
Definition of done	every resolved class emits a new gate or lint; done rule: recurrence rate per class on the debt board falls after the emitter lands, measured over the whole board and never a sample; neg-control: a class with no emitted oracle shows no fall
Matrix cell	PREVENT: every resolved class emits a new gate or lint, measured as recurrence-rate reduction on the debt board · `runtime/nx_failclass.nx` · WATCHING decl
Gate	`nx_refusal_shape_gate` · UNRECORDED — no recorded verdict for this gate (it is declared for the organ but the evidence referee's last run did not record it: it is not on this board's gates map, or the referee has not run since the per-gate record shipped)
Work log	no log row names this rung yet

AD21 Design-choice migration: every adopted design row drives a whole-tree refactoring under the promote rulers · cell WATCHING · gate NO-GATE-DECLARED

Contract	`af_design_migrate` · executor Organ · 3 u · stands on AD9, AD18
Definition of done	OPERATOR 2026-09-22: as the language improves from the first bit up to the last, autodev keeps the estate aligned. For every design row on a board whose status is adopted and whose shape is mechanical (a shared helper replacing clones, a literal replaced by a derivation, a gate shape migrated, a syntax form rewritten), a migration recipe composes the incumbents (nx_oo_extract for consolidation with its byte-identical or behaveprobe proof, nx_magic derive for literals, nx_gate_migrate for gate shapes, nx_srcfmt for surface forms) over the WHOLE tree, never a sample; each run appends a before and after offender census per design id (files touched, files already converged and left byte-identical, files refused with the rule named) and writes the after count into that design row's measured-effect field on its board; the gate proves a planted offender is migrated, an already-converged file is untouched byte-for-byte, and a refused migration leaves the tree unchanged; the census is the ratchet: an adopted design row whose offender count rises is RED on the autodev board. First worklist: lang LD8 (six RIFF writers and the print, length, parse and concatenation clones), LD6 (value-named constants in the G2P and formant organs), LD5 (verdict-last across the gate fleet)
Matrix cell	Design-choice migration keeps the estate converged · `runtime/nx_design_migrate_lib.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_design_migrate_lib`; declare `nx_design_migrate_lib TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	2026-09-22 queue 2026-09-22 declared from the lang board's first design ledger; the migration primitives exist as incumbents (nx_oo_extract promoted, nx_magic derive, nx_gate_migrate); what is missing is the lane that reads design rows, runs them tree-wide and journals the census back to the row. Depends on ecosystem EC66 for the ledger to be readable by one lib 2026-09-22 queue 2026-09-22 MINING LANE from ecosystem EC71 (operator: nothing archived is dropped into the dark): when an artifact is superseded, the migration organs on this board extract every MCP verb, API route, plan step, clock row, ritual and RACI responsibility it carried into rungs on the owning boards before the archive closes, and the mining receipt is what the lifecycle ruler reads; AD21 (design migrate) is the sibling lane and shares the offender census ratchet. The lang board's zero add-on census (LN80: every foreign program a workflow forks, by name and by the language gap it stands in for) is this lane's worklist feeder: each add-on retired by a language rung is an autodev migration with a byte or behaviour proof, never a hand rewrite

dossiers=25 cells_joined=25 no_matrix_row=0 (sum 25) gates_recorded=0 gates_unrecorded=3 no_gate_declared=22 (sum 25 of cells_joined) evgates=UNSTAMPED

Capability comparisons

Compare the field, search individual capabilities and open their source and adoption evidence. Documented presence does not establish comparative quality.

Position map — centrality and distinctiveness

The four-quadrant map the field uses for brand strategy (Dawar and Bagga, HBR June 2015), re-derived from this matrix on every publish. Centrality is the share of the category's feature mass a player covers, each feature weighted by how many hold it; distinctiveness is the average lead over each rival on the rows the player holds; breadth is the depth-weighted share of the whole matrix (the bubble); depth is how deeply the rows held are held; momentum is the day-over-day move off the spine (green rising, red falling, grey until day two); the dashed path runs first day → previous day → today. Dividers are the category means. Axes are fitted to the field of play, so read the tick numerals, not the frame. Rival marks are documented presence, so a rival's position reads the record, never its quality. The picture grades its own readability below; the table beside it is the same data for a screen reader or a second method.

Rows. 41 feature rows mapped of 41 matrix rows; 0 substrate rows (kernels, bounds and instruments the plan's rungrole marks substrate) are the estate's scaffolding, not category features, and stay off the map

Views. 2D cut: centrality, distinctiveness · 3D cube: centrality, distinctiveness, breadth · axes registered: centrality, distinctiveness, breadth, depth, momentum · a board picks its own in knowledge/compare/autodev.cdmap (cut|x|y, cube|x|y|z); absent = the HBR defaults

bubble area = breadth · ring = momentum (green rising, red falling, grey until day two) · dashed = category means · axes fitted to the field of play: centrality 650–900, distinctiveness 0–600 of 0–1000 permil (the full range put every player in one corner)
readability of the 2D cut, self-graded by the layout ruler: label overlaps 0 · labels over marks 0 · off-canvas 0 · unresolved labels 0 · mark overlaps 1 (a fact of the data: two players that close are that close) · data spread 700 permil of the plot · quadrant words unseated 0
text contrast, measured with wcag2-ratio (floors from contrast.conf), light theme: labels 16.24 (floor 4.50) · notes 16.24 (floor 4.50) · quadrant words 5.89 (floor 4.50) · tick numerals 5.89 (floor 4.50) · axis titles 5.59 (floor 4.50) · dark theme: labels 13.78 (floor 4.50) · quadrant words 5.55 (floor 4.50) · axis titles 5.97 (floor 4.50) · dark classes under their floor 0 (one figure serves both themes: a dark shortfall is a token to fix, never a class to hide) · classes refused under their floor 0 (a refused class is not drawn; the scale classes are measured, never hidden) · export: SVG PNG (receipt, rendered by the estate's own rasteriser from this page)

axes fitted: centrality 650–900 · distinctiveness 0–600 · breadth 350–600 permil · farther bubbles are painted first, nearer ones over them · the floor shadow is each bubble's (x, y) at height 0
readability of the 3D cube, self-graded by the layout ruler: label overlaps 0 · labels over marks 0 · off-canvas 2 · unresolved labels 0 · mark overlaps 5 (a fact of the data: two players that close are that close) · data spread 512 permil of the plot

Nishi Claude Code Codex CLI OpenHands Live-SWE-agent
10 panels over 5 registered axes, every pair once (the lower axis on x, the higher on y) · each panel fitted to its own field of play, first and last tick numerals shown · no labels in a small cut, the legend names the colours; the grade below reports the mark terms only (MM summed over the panels, spread averaged)
readability of the small multiples, self-graded by the layout ruler: label overlaps 0 · labels over marks 0 · off-canvas 0 · unresolved labels 0 · mark overlaps 21 (a fact of the data: two players that close are that close) · data spread 543 permil of the plot

Player	Quadrant	centrality	distinctiveness	breadth	depth	momentum	Rows held	Days on spine	First seen
Nishi	Unconventional	675	527	544	827	0 since 2026-09-23	27	10	2026-09-15
Claude Code	Mainstream	833	146	398	777	0 since 2026-09-23	21	10	2026-09-15
Codex CLI	Mainstream	833	126	382	746	0 since 2026-09-23	21	10	2026-09-15
OpenHands	Mainstream	859	128	390	727	0 since 2026-09-23	22	10	2026-09-15
Live-SWE-agent	Mainstream	850	152	398	710	0 since 2026-09-23	23	10	2026-09-15

Day	Matrix rows reviewed	Players recorded
2026-09-15	40	5
2026-09-16	40	5
2026-09-17	40	5
2026-09-18	40	5
2026-09-19	40	5
2026-09-20	40	5
2026-09-21	40	5
2026-09-22	40	5
2026-09-23	41	5
2026-10-08	41	5

How to read this evidence. A Nishi source mark records whether the named symbol is declared in the selected source file. It does not establish working behavior or accepted delivery. A watching tag names an expected source entry; discovering it updates source coverage only. A dark tag means the file exists but the named symbol was not found: inspect the source and contract before assuming the capability is missing or renaming code. Peer codes are authored assessments; unassessed means no comparison is established, not absence. Qualification, deployment identity, runtime observations and acceptance must carry their own evidence. Adoption records the catalog's observed path through source, build, promotion, registration and invocation; libraries use importer evidence and gates use declared execution surfaces. A full catalog path does not establish quality, freshness or a successful user outcome. Partial and unknown states retain their stated next action. Census stamps: importers asof 1790637268, gate census asof 1790104131 (unix seconds; -1 = census absent).

Capability matrix — measured against source

◉ authored stronger claim; source present● present◐ partial○ absent? unassessed; excluded from relative comparison · click any capability for its evidence

Peer cell coverage: authored assessed 164, unassessed 0. Authored assessments are not verified outcomes.

Capability	Nishi	Claude Code	Codex CLI	OpenHands	Live-SWE-agent
INTAKE
the estate's own backlog as executable tasks -- the debt board (4,223 rows, 945 at sev 7 or above) Measured: `db_open_count` exists in `runtime/nx_debt.nx`, verified at emit. The rivals take tasks from GitHub issues and chat: Claude Code routes bug reports from Slack to pull requests and triages issues in CI [claudecode], Live-SWE-agent consumes SWE-bench instances [liveswe]; ours reads a plane whose every row already carries a severity and a scope, and the same board is where a resolved task closes Adoption: `RUN-BY:actlog` — fully adopted (top of its ladder). Pros Implementing source symbol found; runtime acceptance separate; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	●	●	●	●	●
foreign-bug instances admitted only if the bug REPRODUCES on a fresh build (equivalent-mutant discard) Source present; authored stronger claim: `has_failing` in `runtime/nx_autofix_intake_gate.nx`, verified at emit. SWE-bench Verified instances were human-validated [swebench] and SWE-rebench V2 filters unsound instances with an ensemble of LLM judges [swerebenchv2]; no rival HARNESS validates the task it is handed. Ours refuses any instance with no failing per-function row, so a task that cannot fail never enters the resolve rate. PROMOTED and UNREGISTERED today Adoption: `GATE:LIVE trial=RED` — fully adopted (top of its ladder). Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agent; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
LOCALIZE
the failing function found from the grader's own per-function rows, never told Measured: `fn_failing` exists in `runtime/nx_autofix_auto.nx`, verified at emit. The agent-computer-interface lineage localises by navigating the repository and running tests [sweagent]; the ADI adds a function-level Frame Lifetime Trace and lifts existing agents 6.2 to 18.5 percent [adi]. Ours reads FNRES rows from a fresh grader run -- exact, cheap and single-function, which is the whole of its scope today Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run. not adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run Pros Implementing source symbol found; runtime acceptance separateCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run	●	●	●	●	●
PROPOSE
the fix-loop episode -- localise, generate, apply, fresh compile plus run as the sole judge, revert proven exact Measured: `episode` exists in `runtime/nx_autofix_auto.nx`, verified at emit. Every rival column resolves repository-level issues [openhands] [liveswe]; ours is one function at a time with a 1.5B local maker and the binary is BUILT-UNPROMOTED. The mechanism is the same shape as SWE-agent's generate-run-revise loop [sweagent] with a deterministic judge in place of a test suite Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run. dossiernot adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run Pros Implementing source symbol found; runtime acceptance separateCons behind Claude Code (leads), Codex CLI (leads), OpenHands (leads), Live-SWE-agent (leads); not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run	●	◉	◉	◉	◉
sampled retries after both greedy rounds miss (three samples at temperature 0.8, a deterministic seed ladder) Measured: `lf_gen_t` exists in `runtime/nx_autofix_auto.nx`, verified at emit. Test-time compute is where the cheap gains are: tournament voting and distill-refine lifted Claude-4.5-Opus from 70.9 to 77.6 on Verified [ttcagentic], and 32 rollouts with a generative verifier gained up to 7.9 points [swezerohero]. Ours samples exactly three, a literal in source that this board names as debt, and selects only by the judge Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run. not adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run Pros Implementing source symbol found; runtime acceptance separateCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run	●	●	●	●	●
VERIFY
the SWE-bench contract as the sole judge -- FAIL_TO_PASS flips AND PASS_TO_PASS holds Measured: `swb_resolved` exists in `runtime/nx_swebench_local_gate.nx`, verified at emit. The contract is the benchmark's own [swebench]; the gate proves the cheat-proof tooth that a regressive fix is NOT resolved. Rivals are graded by the same contract on the Python set; ours grades a sovereign NishiLang analog and says so on the page. PROMOTED and UNREGISTERED today Adoption: `GATE:LIVE trial=GREEN` — fully adopted (top of its ladder). dossier Pros Implementing source symbol found; runtime acceptance separate; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	●	●	●	●	●
byte-exact deterministic build plus judge -- the verify signal is noise-free Source present; authored stronger claim: `fg_run` in `runtime/nx_forge.nx`, verified at emit. Every rival judges by a test suite, which can flake and which a replay script can beat when verification is not privileged [statprecipice]; a GREEN here is a provable equality. This is the estate's largest structural advantage and every open rung below leans on it Adoption: `LIB-WIRED importers=8 nonval=5` — fully adopted (top of its ladder). Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agent; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	●	●	●	●
non-vacuity by mutation -- can this gate ever fail? the subject is rebuilt and deployed per mutant and restored by hash Source present; authored stronger claim: `gb_try` in `runtime/_hdl_build/nx_gate_bite.nx`, verified at emit. The Darwin Godel Machine validates each self-modification empirically on a benchmark [dgm] but never asks whether the benchmark can fail; no rival harness mutates its own oracle. Ours does, and the failure-record ruler below found nine gates that have never once gone red Adoption: `RUN-BY:actlog` — fully adopted (top of its ladder). dossier Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agent; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
the promote ruler -- a capability-loss diff naming every lost run before a candidate replaces the live binary Source present; authored stronger claim: `cd_show_lost` in `runtime/nx_contentdiff.nx`, verified at emit. Regression prevention is a named open challenge of the harness field [codeharness]; ours refuses a promote that loses a run and prints the run. Its known blindness -- a code-only delta with no literal loss -- is why the verify chain below adds behaviour and size legs beside it Adoption: `RUN-BY:fork:nx_rebuild_drain` — fully adopted (top of its ladder). Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agent; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
SHIP
the ship loop as one organ -- build, prove by gate, contentdiff, behaveprobe, promote, journal with the sha Measured: `os_jrnl` exists in `runtime/nx_organ_ship.nx`, verified at emit. Claude Code stages changes, writes commits and opens pull requests [claudecode]; Codex CLI runs locally on the developer's machine [codex]; OpenHands ships patches from a sandbox [openhands]; Live-SWE-agent emits a patch per instance [liveswe]. Ours ships to a serving root with a banked rollback and journals every stage; the rivals' git-native flow is the better product surface and the reason the cell reads Best for them Adoption: `RUN-BY:clock` — fully adopted (top of its ladder). Pros Implementing source symbol found; runtime acceptance separate; ahead of Live-SWE-agent; fully adopted on the estate ladderCons behind Claude Code (leads), Codex CLI (leads)	●	◉	◉	●	○
GUARD
patch-cascade tooth -- a second failed fix refuses a third (rule 3 mechanised, the maximum a conf row) Source present; authored stronger claim: `PG_DEFMAX` in `runtime/_hdl_build/nx_patch_guard.nx`, verified at emit. Agents average 8.8 test runs per task and range from 2 to 19 [runornot]; none of the four refuses a cascade by construction. Registered and dark today -- the tooth exists and the loop does not call it yet Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). not adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked) Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agentCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked)	◉	○	○	○	○
the egress warden -- terms ledger plus classify-at-the-door before any byte reaches an external model Source present; authored stronger claim: `xl_egress_class` in `runtime/nx_extllm.nx`, verified at emit. The rival products send the repository to the model by design [claudecode] [codex]. Ours refuses any prompt byte not on the operator's public-release manifest, because the ox-alpha page says prompts are retained and not used for training [oxalpha] while the contract that binds it says they are collected for training [stealtheula] -- the contract governs, and the ledger row carries its pin Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). not adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked) Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agentCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked)	◉	○	○	○	○
SAMPLER
an external frontier sampler behind the warden -- ox-alpha, 1,048,576-token context, free Measured: `xl_call` exists in `runtime/nx_extllm_call.nx`, verified at emit. The model page describes a reasoning model for coding and sustained agentic work [oxalpha]; the call organ ledgers provider, model, prompt hash, bytes, class, terms pin and verdict per call and attributes every 429 to its source. Capability facts only on this page, by the EULA [stealtheula] Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). dossiernot adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked) Pros Implementing source symbol found; runtime acceptance separateCons behind Claude Code (leads), Codex CLI (leads); not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked)	●	◉	◉	●	●
the local sovereign no-float Coder tier (0.5B and 1.5B) loaded bits-up Source present; authored stronger claim: `load_named_q16` in `runtime/nx_nofloat_llm.nx`, verified at emit. No rival runs a model through its own integer stack; this is the sovereign bottom rung of the sampler ladder and the tier the July resolve rate was measured on. Scale is the capability lever the record already measured Adoption: `LIB-WIRED importers=34 nonval=8` — fully adopted (top of its ladder). Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agent; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
CONTEXT
verified context pack emitted from sources, regenerable Measured: `fc_emit` exists in `runtime/nx_forge_ctx.nx`, verified at emit. A harness is the code that decides what to store, retrieve and present to the model [metaharness] and its component list is now standard across the field [lilharness]; Claude Code reads CLAUDE.md and builds auto memory [claudecode]. Ours is vetted from sources with every rule witnessed and auto-retired when the compiler heals Adoption: `LIB-WIRED importers=7 nonval=4` — fully adopted (top of its ladder). Pros Implementing source symbol found; runtime acceptance separate; fully adopted on the estate ladderCons behind Claude Code (leads), Codex CLI (leads)	●	◉	◉	●	●
sovereign semantic code retrieval (jina recipe, integer embedding path) Measured: `ci_query_top1` exists in `runtime/nx_code_index.nx`, verified at emit. Base-weight quality, 3 of 6 top-1, a ratchet; the coding domain owns the contrastive tune that closes it. Here it is the retrieval leg of the localiser rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them Adoption: `LIB-WIRED importers=4 nonval=2` — fully adopted (top of its ladder). Pros Implementing source symbol found; runtime acceptance separate; fully adopted on the estate ladderCons behind Claude Code (leads)	●	◉	●	●	●
FAILURE RECORD: 880 verdict streams classified against a published taxonomy; nine gates that have fired GREEN on every run of their history named as a bite worklist Source present; authored stronger claim: `fc_alert_score` in `runtime/nx_failclass.nx`, verified at emit. HAL inspected 2.5B tokens of agent logs with an LLM to find behaviours nobody reported, including agents searching for the benchmark instead of solving it [hal]; ours classifies the whole record deterministically and prints a partition that sums. A perfect history is the absence of evidence, and this ruler is how the loop chooses which oracles to bite first Adoption: `RUN-BY:clock` — fully adopted (top of its ladder). Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agent; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
EVOLVE
the evolution harness with an external fitness (a gate permil) and a known-optimum control Measured: `ev_fit_ext` exists in `runtime/_hdl_build/nx_evolve.nx`, verified at emit. AlphaEvolve's loop is an evaluator feeding an evolutionary search [alphaevolve]; the Darwin Godel Machine keeps an archive of agents validated on benchmarks [dgm]; Live-SWE-agent rewrites its scaffold on failure [liveswe]. Ours validates the machinery on a fitness with a known optimum before trusting it on a gate, and it is PROMOTED and UNREGISTERED Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). not adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked) Pros Implementing source symbol found; runtime acceptance separate; ahead of Claude Code, Codex CLI, OpenHandsCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked)	●	○	○	○	●
GRADE
the SWE-bench Verified grader owning its denominator -- python is the oracle, never the scorer Measured: `sg_verdicts` exists in `runtime/_hdl_build/nx_swebv_grade.nx`, verified at emit. The Verified leaderboard's numbers are computed by the benchmark harness and dominated by industry entries on proprietary models [swebenchcase]; ours computes the number itself over a results plane. Registered and unauthorised today Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). not adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked) Pros Implementing source symbol found; runtime acceptance separate; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agentCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked)	●	○	○	○	○
HEAL
the heal plane -- service status and the remediation log the loop closes against Measured: `nh_metrics` exists in `runtime/nx_heal.nx`, verified at emit. No rival has a notion of the estate it runs in; a resolved runtime defect here is one whose heal-plane row goes quiet. Registered and dark today Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). not adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked) Pros Implementing source symbol found; runtime acceptance separate; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agentCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked)	●	○	○	○	○
RIGOR
every published number carries n, a Wilson interval with hierarchical bootstrap over the nested task structure, seed, harness hash and model pin -- and the emit REFUSES a bare percentage Source present; authored stronger claim: `gv_envelope` in `runtime/nx_gate_verdict.nx`, verified at emit. Posterior means with credible intervals and a declare-a-difference-only-on-non-overlap rule replace pass at k [dontpassk]; naive aggregation over correlated tasks under-covers badly and Wilson plus hierarchical bootstrap corrects it [statprecipice]. Today our autograde page prints 92 percent with no n. The envelope lives in the gate base class so that /compare/gen's referee run (G5) and this board publish under ONE ruler Adoption: `LIB-WIRED importers=2826 nonval=854` — fully adopted (top of its ladder). dossier Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agent; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
harness disclosure manifest -- tools, prompts, context builder, verifier, budgets and sampler tier, hashed and printed beside every result Source present; authored stronger claim: `os_harness_manifest` in `runtime/nx_organ_ship.nx`, verified at emit. Harness configuration governs more variance than model choice and undisclosed harnesses make leaderboards misleading [harnessthesis]; OpenHands and Live-SWE-agent publish their code [openhands] [liveswe] but no per-result manifest, and the two products publish neither. Ours prints the hash the result was produced under, so two numbers are comparable only when their manifests match Adoption: `RUN-BY:clock` — fully adopted (top of its ladder). dossier Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agent; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	◐	◐
INTAKE
one task plane from every board -- debt, gate roster RED, failclass LATCHED, unwired, magic, artifactdrift BEHIND -- each task carrying its executable oracle and a RED-before receipt, partition summing Source present; authored stronger claim: `af_intake_board` in `runtime/nx_autofix_auto.nx`, verified at emit. The estate's boards hold thousands of tasks with machine oracles already attached, which no rival's intake can claim. ChainSWE shows that a stream of related defects in one codebase is the hard and realistic setting, with up to a 70 percent drop as chains lengthen [chainswe]; the plane keeps the chain order Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run. dossiernot adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agentCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run	◉	○	○	○	○
LOCALIZE
retrieval-backed localisation -- code index plus failclass plus gate stderr -- with hit at 1 measured on held-out tasks DARK — `runtime/nx_autofix_auto.nx` EXISTS but does not declare `af_localize_index`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. Dynamic function-level evidence beside static retrieval is what the ADI measured as the cheapest lift [adi]; the contract publishes hit at 1 with its interval, never a bare rate dossierwatching af_localize_index Pros none measured yetCons behind Claude Code, Codex CLI, OpenHands, Live-SWE-agent; open contract; expected source declaration not established	○	●	●	●	●
SAMPLER
sampler tiers as DATA -- local 1.5B, ox-alpha through the warden, Claude for judged work -- selected per task class by a conf row, never by code DARK — `runtime/nx_autofix_auto.nx` EXISTS but does not declare `af_sampler_tier`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. OpenHands runs many model backends behind one agent [openhands]; ours adds the clearance: the free tier sees only U bytes [stealtheula], the local tier sees everything, and the Claude tier is the escalation whose output may never grow the sovereign corpus dossierwatching af_sampler_tier Pros none measured yetCons behind Claude Code, Codex CLI, OpenHands (leads), Live-SWE-agent; open contract; expected source declaration not established	○	●	●	◉	●
REPRO
a bug-reproduction tooth cogenerated with every fix -- a neg-control that fails before and passes after, and the planted mutant must die DARK — `runtime/nx_autofix_auto.nx` EXISTS but does not declare `af_repro_tooth`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. Google's study of 120 bugs found cogenerating the reproduction test in the same patch costs no fix rate and beats a separate test agent [brtcogen]. Ours makes the tooth a gate row, so the loop leaves prevention behind it every time it fixes something dossierwatching af_repro_tooth Pros none measured yetCons open contract; expected source declaration not established	○	○	○	○	○
VERIFY
the verify chain as one pipeline -- build, gate, bite, contentdiff, behaveprobe, magic and unwired ratchets -- with an execution budget measured per task class DARK — `runtime/nx_autofix_auto.nx` EXISTS but does not declare `af_verify_chain`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. Prohibiting execution cost only 1.25 points of resolve rate across 3,000 attempts and the benefit concentrated on a subset of instances [runornot]; the budget is a conf row derived from the measured benefit per class, and the ratchets make a fix that adds a magic number or an unwired function REFUSE dossierwatching af_verify_chain Pros none measured yetCons behind Claude Code, Codex CLI, OpenHands, Live-SWE-agent; open contract; expected source declaration not established	○	●	●	●	●
PROPOSE
best-of-N against the free verifier DARK — `runtime/nx_forge.nx` EXISTS but does not declare `fg_best_of_n`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. The same contract coding names as CD2 and extllm as XL5, so three boards flip on one organ; pass at N monotone in N and NONE reported when nothing is GREEN [ttcagentic] [swezerohero] dossierwatching fg_best_of_n Pros none measured yetCons behind Claude Code, Codex CLI, OpenHands, Live-SWE-agent; open contract; expected source declaration not established	○	●	●	●	●
multi-file coordinated patch set, the judge builds the whole closure DARK — `runtime/nx_forge.nx` EXISTS but does not declare `fg_multi_file`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. SWE-bench Pro tasks take a professional hours to days and patch multiple files [swebenchpro]; the same contract coding names as CD5 -- our loop is single-function until it lands dossierwatching fg_multi_file Pros none measured yetCons behind Claude Code (leads), Codex CLI (leads), OpenHands (leads), Live-SWE-agent (leads); open contract; expected source declaration not established	○	◉	◉	◉	◉
CONTROL
null resolvers -- empty patch, revert patch, verbatim echo, replay of a prior solution -- must score zero, printed beside every run Source present; authored stronger claim: `af_null_control` in `runtime/nx_autofix_auto.nx`, verified at emit. A replay script that never observes the screen beat frontier models on static benchmarks, with expected success exactly the source agent's pass at k [statprecipice]; the echo class already cost this loop three misses in July, so the null controls are the board's own history made mechanical Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run. dossiernot adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agentCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run	◉	○	○	○	○
GUARD
untrusted-input admission -- no external bug, patch or trajectory enters the tree or a training set unless it is DATA-ONLY, provenance-pinned and maintainer-merged; deny by default Source present; authored stronger claim: `af_admit_untrusted` in `runtime/nx_autofix_auto.nx`, verified at emit. Two sev-8 debt rows (1784661149, 1784661346) demand this before any ingestion or training. HAL found agents that searched HuggingFace for the benchmark rather than solving it [hal]; a poisoned bug-fix pair is the same class from the other side. Neg-control: a planted unmerged patch must be refused by name Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run. dossiernot adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run Pros Source symbol found; stronger capability claim is authored; ahead of Claude Code, Codex CLI, OpenHands, Live-SWE-agentCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run	◉	○	○	○	○
SANDBOX
every candidate builds and runs in an isolated scratch root with time and resource limits, never the serving root Measured: `af_sandbox_root` exists in `runtime/nx_autofix_auto.nx`, verified at emit. OpenHands executes in sandboxed environments by design [openhands]; the DGM ran every self-modification under sandboxing and human oversight [dgm]; the Codex README as mirrored says nothing about sandboxing so its cell is coded from silence [codex]. Ours today edits in place with a banked copy and a proven revert, which is a rollback and not an isolation Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run. dossiernot adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run Pros Implementing source symbol found; runtime acceptance separate; ahead of Codex CLICons behind OpenHands (leads); not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run	●	●	◐	◉	●
LEDGER
one row per episode -- task, tier, samples, executions, verdict, tokens, wall, cost -- and the page reads the plane, never a hand number Measured: `af_episode_ledger` exists in `runtime/nx_autofix_auto.nx`, verified at emit. HAL released 2.5B tokens of logs so that behaviour could be audited [hal]; ours writes the row the rigor envelope aggregates and the cost the engine-shift gauge reads Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run. dossiernot adopted: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run Pros Implementing source symbol found; runtime acceptance separateCons not adopted yet: callable, authorised, completed MCP invocation not established by supplied journal window; lifetime history unknown (direct execution may be recorded elsewhere; this scan does not establish lifetime use); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by unpromoted-source -- a mention, not a run	●	●	●	●	●
REFEREE
SWE-rebench V2 and SWE-bench Pro post-cutoff multilingual subset, run bench-only in an isolated sandbox host, graded by our grader DARK — `runtime/_hdl_build/nx_swebv_grade.nx` EXISTS but does not declare `sg_rebench_run`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. 32,079 executable tasks across 20 languages with pre-built images [swerebenchv2] and a held-out long-horizon set [swebenchpro] are the referees; Verified is saturated and industry-skewed [swebenchcase], so it stays a sanity check. The run reports n, the interval and the harness hash or it does not publish dossierwatching sg_rebench_run Pros none measured yetCons behind Claude Code (leads), Codex CLI (leads), OpenHands (leads), Live-SWE-agent (leads); open contract; expected source declaration not established	○	◉	◉	◉	◉
the Aider-Polyglot runner DARK — `runtime/nx_forge.nx` EXISTS but does not declare `fg_polyglot_bench`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. The same contract coding names as CD6; the first number is whatever it is rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them dossierwatching fg_polyglot_bench Pros none measured yetCons behind Claude Code (leads), Codex CLI (leads), OpenHands (leads), Live-SWE-agent (leads); open contract; expected source declaration not established	○	◉	◉	◉	◉
EVOLVE
harness evolution as PROPOSALS -- harness parameters are conf rows, the loop proposes edits scored on a HELD-OUT split against a matched-budget test-time-scaling baseline, and nothing auto-admits DARK — `runtime/_hdl_build/nx_evolve.nx` EXISTS but does not declare `ev_harness_search`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. Meta-Harness and AHE automate the search [metaharness] [ahe] and the essay names it the near-term path to self-improvement [lilharness]; the July-2026 re-evaluation found evolved harnesses rarely beat simple test-time scaling at matched budget and generalise poorly [harnessevalrethink]. That finding is the accept rule: a proposal ships only if it beats the matched-budget baseline on held-out tasks with non-overlapping intervals dossierwatching ev_harness_search Pros none measured yetCons behind Live-SWE-agent (leads); open contract; expected source declaration not established	○	○	○	○	◉
MEMORY
bounded failure memory -- a per-class failure library with eviction, feeding prevention rather than prompts DARK — `runtime/nx_failclass.nx` EXISTS but does not declare `fc_failmem`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. Claude Code builds auto memory as it works [claudecode]; Live-SWE-agent keeps its original context to prevent drift while evolving [liveswe]. Ours bounds growth by construction because a store that outgrows its reader fails silently and tail-first dossierwatching fc_failmem Pros none measured yetCons behind Claude Code, Codex CLI, OpenHands, Live-SWE-agent; open contract; expected source declaration not established	○	●	●	●	●
CHAIN
sequential dependent tasks on one organ family without a reset, resolve rate published per chain length DARK — `runtime/nx_autofix_auto.nx` EXISTS but does not declare `af_chain_run`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. ChainSWE is the first benchmark of this setting and reports the drop with chain length [chainswe]; our debt board already holds chains (the same defect filed five times in three weeks), so the number is ours to publish first dossierwatching af_chain_run Pros none measured yetCons open contract; expected source declaration not established	○	○	○	○	○
DISTILL
terms-gated harvest of verified episodes into the sovereign corpus -- never from ox-alpha or Claude DARK — `runtime/nx_extllm.nx` EXISTS but does not declare `xl_distill_corpus`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. The same contract extllm names as XL9. SWE-HERO's recipe used 300,000 execution-free and 13,000 execution-based trajectories [swezerohero] and SWE-rebench V2 exists to feed RL [swerebenchv2]; our corpus grows only from episodes the byte-exact judge passed and only from providers whose terms permit it, and with every external sampler disabled it must still strictly grow dossierwatching xl_distill_corpus Pros none measured yetCons open contract; expected source declaration not established	○	○	○	○	○
PREVENT
every resolved class emits a new gate or lint, measured as recurrence-rate reduction on the debt board DARK — `runtime/nx_failclass.nx` EXISTS but does not declare `fc_prevent_emit`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. Prevention outranks self-healing outranks post-analysis, and post-analysis exists to build new prevention; no rival column emits an oracle after a fix. The measurement is the board's own recurrence count per defect class before and after dossierwatching fc_prevent_emit Pros none measured yetCons open contract; expected source declaration not established	○	○	○	○	○
Design-choice migration keeps the estate converged Open — watching `runtime/nx_design_migrate_lib.nx` : `af_design_migrate`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. WATCH AD21. Every adopted design row on a board drives a whole-tree migration through the incumbent refactoring organs under the promote rulers, with a before and after offender census per design id as the ratchet; Claude Code, Codex CLI, OpenHands and Live-SWE-agent all refactor on request and none carries a board-declared ledger whose adopted rows drive the migration and count the offenders, coded 3 rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them dossierwatching af_design_migrate Pros none measured yetCons behind Claude Code, Codex CLI, OpenHands, Live-SWE-agent; open contract; expected source declaration not established	○	◐	◐	◐	◐

Rival-claim provenance. Every Yes, Best or Part mark in a rival column is a claim about someone else's product. A cited row carries a reference mark that resolves to a pinned mirror (the refs gate measures that); an uncited row is an observation read and is badged in its evidence until a reference lands. The badge is the receipt, never the proof: a mark proves a mirror exists, not that the mirror supports the code. The rows are counted below.

Rival claims, counted
Class	Count	Note
rows with rival marks	24	rows carrying a Yes, Best or Part in a rival column
cited	21	a reference mark resolves to a pinned mirror
uncited	3	an observation read, badged until a reference lands
rows, summed	24	RECONCILED: the parts sum to the total 24

Delivery and evidence

Inspect risks, technical debt, rendered observations, experiments and references. Read scope and limitations alongside every result.

Release history and work log

The worked plan, on the board. Every leg appends what it measured, landed, retracted, learned and left queued, per rung, so a crash leaves the next seat a ledger here rather than a transcript to mine. Kinds: measure, land, retract, lesson, queue. Newest last.

When	Rung	Kind	Entry
2026-09-02	AD6	measure	2026-09-02: nx_autofix_intake_gate read passed 3/5 verdict=RED on the NAS (15 proposed, 15 NO FNRES ROWS -> DISCARDED, kept=0) and the domain sat on the CLAIMED-BUT-RED list that blocks every deploy. Root cause is not the intake: every sandbox compile goes through _offc/nx_sov_build_run.elf, which exits SBR_ADMIT_REFUSED=6 under the I/O storm, so the grader capture was empty because nothing was BUILT, and the gate read an unbuilt instance as a non-reproducing one. Same class as the sclass dispatch gate and the referee deadline fixed earlier today
2026-09-02	AD6	queue	Source fix landed (10607 B): AIG_BUILDER_ADMIT_REFUSED=6 sorts a refused compile into an unexamined bucket (never discarded, printed BUILDER-ADMISSION-REFUSED), reached = kept+discarded >= 1 is a gv_need precondition, T2/T3/T5 are asserted only when reached, the partition line carries unexamined= and still sums to proposed. Expected on a saturated box: SKIP exit 3 with T1 and T4 still run; expected on a calm box: unchanged 5/5. Build queued behind admission (governor QUEUE at procs_running=8/8; /api/build enqueued); promote, gate_run and the autodev referee re-run owed

Risk register

Risk	Likelihood x impact	Mitigation
A harness that wins on the split it searched and loses on held-out tasks reads as progress	probable x high	AD16 pre-declares the held-out matched-budget accept rule from the July-2026 re-evaluation [harnessevalrethink] and journals every refusal
A stub carrying a contract symbol flips a watch cell green with no capability behind it	possible x high	the flip is the receipt, the rung gate is the proof; every AD rung names its neg-control
The resolver games its own oracle -- edits a gate to pass, or a fix that only satisfies the tooth	possible x high	AD8 requires the planted mutant to die under nx_gate_bite, AD9 refuses gate edits outside the fix's declared scope, AD4 keeps null controls on every batch [statprecipice]
A free tier trains on what it is shown and a probe number leaks to a public page	probable x high	the warden classifies every byte at the door and the publish guard keeps ox-alpha numbers at class C [stealtheula]; this page carries capability facts only
A poisoned external bug-fix pair enters the corpus or the tree	possible x high	AD6 deny-by-default admission, data-only, provenance-pinned, maintainer-merged; neg-control planted and bite-proven before any ingestion
The number keeps changing and the reader learns to distrust the instrument	probable x medium	AD1 publishes the method with every count and AD2 the manifest hash, so a moved number names what moved
An interrupted run yields nothing	probable x medium	AD5 appends each episode as decided; a run interrupted at 95 percent keeps 95 percent

Debt register

Id	Sev	What it is	Unblock
autodev-loop-unpromotednot in this scope	7	nx_autofix_auto is BUILT and UNPROMOTED (325,366 B staged) and nx_autofix_intake_gate, nx_swebench_local_gate and nx_evolve are PROMOTED and UNREGISTERED; the flagship loop of this board cannot be called from the MCP surface, which is the built-and-unwired class the operator named the deliverable.	Promote through nx_organ_ship with the gate named in organ_gate.conf, register the three via /api/tools/register, mint a cap, invoke once, and re-read nx_catalog until all four read LIVE.
autograde-no-denominatornot in this scope	6	/compare/autograde/api.json publishes 92 percent resolved with no n, no interval and no run date beyond generated_unix; the July record shows n was 14, and a percentage without a denominator is the shape every rigor law on this estate forbids.	AD1: the row is emitted from the episode ledger through gv_envelope or it reads REFUSED; until then the page says sovereign analog and nothing about confidence.
autofix-bon-literalnot in this scope	4	AF_BON_N, AF_BON_TEMP_PM, AF_BON_TOPP_PM and AF_BON_TOPK are literals in nx_autofix_auto.nx; the sampler shape is code, not data, so the best-of-N rung cannot be tuned without a rebuild.	AD12 reads them from a conf row; the literals retire with a byte-identical-behaviour proof on the July manifest.
autodev-gates-unrunnable-from-rootnot in this scope	6	MEASURED 2026-08-27 by nx_swcompare_evidence over autodev.gates: nx_swebench_local_gate exits 3 SKIP pass 0 of 0 from the serving root because the fix-loop ledger it reads lives under a home directory (nx_stage/autofix_ledger.log), and nx_autofix_intake_gate exits 1 RED pass 0 of 0 with no proposal rows to admit -- so the two promoted gates of the loop cannot serve as executed evidence for this page today and the evidence stamp reads RED, which this board publishes rather than hides.	AD5 moves the ledger to a plane under knowledge/ that the gate reads by an estate-relative path, and AD3 gives the intake gate a standing proposal plane; both gates then run GREEN from the root and get bitten (C4 never-bitten today), and only then does the stamp turn.
autodev-dark-guardsnot in this scope	5	nx_patch_guard, nx_gated_edit, nx_editjudge, nx_heal, nx_triage and nx_dr_run are registered and dark; the loop has guards it does not call.	AD9 composes the patch guard and the gated editor into the verify chain; invocation is then measured by the actlog rather than asserted.

Release history continued

When	Rung	Kind	Entry
2026-09-03	AD1	land	gv_bare_rate WIRED into gv_verdict, the emitter every gate calls. buildroot/runtime/nx_gate_verdict.nx 35612 -> 38635 B, sha256 799e453ad2781a8b -> 6c6a488fac255b0a. New func gv_note_bare_rate scans the note BEFORE any verdict byte is emitted, so the announcement can never displace the verdict line from the tail where gv_last_line anchors. SILENT WHEN CLEAN by construction, so stdout stays byte-identical for every gate without the defect and no judge-equivalence proof is invalidated. It ANNOUNCES rather than refuses, chosen not conceded: gv_verdict's return value IS the fleet verdict, so a refusing base class would turn every gate whose note quotes a rate RED in a single edit.
2026-09-03	AD1	measure	Wiring proven on the ARTIFACT, never the receipt. /api/build BUILT 57840 with src_path naming nx_rigor_envelope_gate and src_stable=1; nx_contentdiff lost_from_live=0 lost_permil=0 symmetry=ONE-WAY-GAIN verdict=GREEN 3 of 3, its GAINED list naming the new string NX-RIGOR bare-rate-in-note byte_offset= so the code is provably IN the binary; nx_behaveprobe first_diff_off=-1 verdict=IDENTICAL evidence=STRONG; PROMOTED sha256 b292fcf0be57845b prev=1; /api/gate_run against the LIVE promoted binary exit_code=0 verdict=GREEN capture_truncated=0.
2026-09-03	AD1	measure	Adoption census taken BEFORE the edit, coverage_complete=1 corpus_complete=1 over files=23592: gv_bare_rate matches=8, every one its own definition, its own comment, or nx_rigor_envelope_gate which is its own gate; gv_envelope_check matches=1, its definition alone. The refusing half of AD1 was built, gate-proven, and unreachable from every production emitter in the estate.
2026-09-03	AD1	lesson	A RULER PLACED IN THE BASE CLASS IS NOT THEREBY ADOPTED. nx_gate_verdict's own rigor header argues that a ruler in a sibling lib is adopted at advice rates while a ruler in the base class is adopted by every gate that imports it -- and gv_bare_rate then sat in that very base class with ZERO callers. The adoption argument was made and the wiring was not done.
2026-09-03	AD1	retract	Two recommendations withdrawn, the second by measurement taken this session. The first, build nx_autodev_qa because no adjudicator exists, was already withdrawn as a top-8 ranked search read as a population. Its replacement is ALSO refuted: nx_envelope_audit is NOT the adjudicator, not even partly. Reading its source shows a SOURCE-TEXT scanner over the registered tool_allowlist classifying rows DECLARING vs SILENT-CAP-CAND vs UNBOUNDED-OR-NA vs NAME-PROBE-MISS for the SCALE envelope, meaning bounded reads, _CAP and _MAX consts and silent truncation, and its own header states that a detector reading source text cannot enforce a contract defined on runtime output. The rank-2 hit was a HOMONYM: envelope means bounded-read-declaration in nx_envelope_audit and rate-plus-denominator-plus-interval in gv_envelope_check. Two organs, one word, unrelated subjects.
2026-09-03	AD5	land	QA admission contract landed as DATA at knowledge/autodev_admit.conf, 7909 B sha256 6c861099c41ad4cd. Nine NAMED states -- ADMIT, four REFUSE and four UNPROVEN -- each naming its DECIDER organ and the PASSTHROUGH field whose bytes are copied verbatim into the record, because a hand-copied paraphrase of another organ's verdict is a duplicate ruler made of prose. Every UNPROVEN state ABSTAINS and can never grant an ADMIT. UNCONSUMED until an organ names it, and the file says so rather than claiming enforcement it does not have.
2026-09-03	AD5	land	First work queue landed as DATA at knowledge/autodev_workqueue.conf, 7664 B sha256 14c8d96ea4c7b2fa. Three admitted rows, each carrying the machine oracle that decides it and the measurement that put it there; four exclusion rows, each carrying its refusal state and evidence. A SEED, NOT A CENSUS, and it says so.
2026-09-03	AD5	lesson	nx_organ_ship prints prove=NO-GATE-FOUND UNPROVEN -- a NAMED absence, NOT a pass -- at nx_organ_ship.nx:564 and :565 and carries it to the final line at :794, AND PROMOTES ANYWAY. Correct for a human ship and catastrophic for an autonomous worker, because the difference is not the message but who is left to act on it. Three organs exist because this already happened: nx_raidwatch_gate, nx_search_census_gate, and nx_seat_gate, the last recording that a 29-tooth gate sat ONE ARGUMENT AWAY while the ship reported nothing proven.
2026-09-03	AD5	queue	Next rung is the adjudicator that READS knowledge/autodev_admit.conf, plus the census that turns the three-row seed queue into a population. Neither is claimed here.
2026-09-12	AD9	operator	operator-unstall-20260912: Resume Nishi autodevelopment through existing nx_autofix_auto and nx_autofix_lib with a reusable MCP/API workflow, alongside Beach, Browser, Video and Search. Fresh intake job 1789237042 succeeds: total=694 gates_red=155 drift=38 magic=501 debt_refused_no_oracle=4450 partition=RECONCILES. CURRENT BLOCKER job 1789237101: all 15 existing proposal sandbox runs exit 7 REFUSED-TREE-CANON INPUT, compiler not launched, yet intake counts discarded=15/unexamined=0 and returns 4/6 RED. This supersedes the old load-exit-6 diagnosis. Smallest repair: carry validated canonical build context into the existing sandbox via the build owner's supported route and classify unexecuted/refused runs as UNEXAMINED with exact reason. Acceptance: same 15-proposal intake builds in isolation, reproducing bugs retained, equivalent mutant discarded, missing canon abstains, tree unchanged, named gate and no-provider-call receipt; then exercise existing admission contract before any autonomous promotion. Registration and intake invocation do not establish an adopted repair loop.
2026-09-22	AD21	queue	2026-09-22 declared from the lang board's first design ledger; the migration primitives exist as incumbents (nx_oo_extract promoted, nx_magic derive, nx_gate_migrate); what is missing is the lane that reads design rows, runs them tree-wide and journals the census back to the row. Depends on ecosystem EC66 for the ledger to be readable by one lib
2026-09-22	AD21	queue	2026-09-22 MINING LANE from ecosystem EC71 (operator: nothing archived is dropped into the dark): when an artifact is superseded, the migration organs on this board extract every MCP verb, API route, plan step, clock row, ritual and RACI responsibility it carried into rungs on the owning boards before the archive closes, and the mining receipt is what the lifecycle ruler reads; AD21 (design migrate) is the sibling lane and shares the offender census ratchet. The lang board's zero add-on census (LN80: every foreign program a workflow forks, by name and by the language gap it stands in for) is this lane's worklist feeder: each add-on retired by a language rung is an autodev migration with a byte or behaviour proof, never a hand rewrite
2026-09-23	AD16	operator	Estate-wide continuous improvement must deliver both an inspectable human outcome and an accountable machine feedback loop. Native research captures retained and hashes checked for DGMv2 and the independent research case-study evaluation2607.27191. Candidate and evaluator changes require separate review; recovery must be exercised. No achieved superiority, autonomous scientist or positive swarmROI is claimed. Existing loadgov LV21/LV25, PM intake1790132172 and shared publication owners are the initial integration path, not a duplicate service.
2026-09-23	AD16	queue	INTAKE 1790132172 routed by nx_intake_route (nishi-review-via-codex, epoch 1790146428) from operator-via-codex: Estate-wide continuous improvement: pair an inspectable human outcome with an accountable recurring machine loop. Compare intake through native experiments, independent evaluation, guarded release, re -- sponsor intake-1790132172 plus1

On these two registers. Rows are declared in the domain's plan file and carry the debt id, which is the join key back to the sovereign debt plane — that plane, not this page, is the authority on state. Reconciling them automatically (the regen reading the plane and refreshing these rows) is a named, owed rung; until it lands, treat an id here as a pointer to look up, not a status to trust.

Honest verdict. Honest position, measured 2026-08-27. The pieces of an autonomous resolver already exist here and most of them are dark. The fix loop (nx_autofix_auto: self-localising from the grader's own per-function rows, a local 1.5B maker, fresh compile plus run as the sole judge, revert proven exact, sampled retries after greedy misses) is BUILT and UNPROMOTED; its intake gate and its SWE-bench-contract grader are PROMOTED and UNREGISTERED; the evolution harness, the patch-cascade guard, the gated editor and the heal plane are registered and dark. The published scoreboard says 92 percent of our foreign-bug instances resolve under the SWE-bench contract [swebench] and prints no denominator -- the first thing this board's rigor rung refuses is our own page. What the estate genuinely has that the field does not is the ORACLE: a deterministic byte-exact judge so the verify signal is noise-free, 2,317 gate sources that are executable specifications, a mutation harness that asks whether a gate can fail at all, and a promote ruler that names every lost run before a binary goes live -- regression prevention, which the harness survey lists as an open challenge [codeharness]. What the field has that we do not is repository scale and a public number: the rival columns resolve real multi-file issues [sweagent] [openhands] [liveswe] while ours is single-function scope with no external-leaderboard result. The August-2026 record sets the design: harness configuration explains more variance than model choice and must be disclosed [harnessthesis]; test-time compute buys the largest cheap gains when a verifier exists [ttcagentic] [swezerohero]; execution should be budgeted, not defaulted [runornot]; the reproduction test belongs in the patch [brtcogen]; harness self-evolution must be scored on held-out tasks against a matched-budget baseline or it is overfitting wearing a win [harnessevalrethink] [metaharness] [ahe]; Verified is saturated and sociologically skewed [swebenchcase], so the referee is SWE-rebench V2's post-cutoff multilingual pool [swerebenchv2] and SWE-bench Pro's held-out set [swebenchpro], with chained maintenance as the honest next axis [chainswe]; and every number carries an interval that respects the nested structure of agent benchmarks [dontpassk] [statprecipice] and ships beside a null control that a replay script cannot beat. The free frontier sampler is ox-alpha with a 1,048,576-token context [oxalpha], reached only through the egress warden because the contract that binds it collects prompts for training [stealtheula]. Nothing external is trusted: the debt board carries two sev-8 rows demanding deny-by-default admission of foreign bugs and fixes before any ingestion or training, and this board makes that a rung with a neg-control rather than a promise.

Person · product · place — not yet measured for this domain

Every compare carries this layer. Declare knowledge/compare/autodev.ppp (rows surface|nishi or c1..c4|label|url|connect naming OUR live surface and each rival's front door), run nx_ppp_probe domain autodev, and this section fills itself on the next beat: the same ruler on both sides — privacy and CX (third-party hosts, tracker classes, cookies, security headers), design and longevity (design hygiene, computed WCAG contrast, render-blocking resources, unsized media, script weight, theme and motion queries), findability (landmarks, skip link, on-site search, breadcrumb, headings, internal links).

The field — discovered, not chosen

Rows written by nx_field_discover from autodev.seeds: the industry's own lists (Wikipedia wikitext, GitHub topics, awesome lists) read mechanically, every candidate counted across seeds. The matrix columns above are a SEAT'S pick; this band is the population they were picked from, and the stats line measures one against the other. A rival here is a lead, never a verdict — it earns a column when its capabilities are read and pinned.

Discovery receipt. One row per counter the run wrote, in the receipt's own order; the rival rows follow in their own table.

Field discovery, counted
Class	Count	Note
seeds	3
scanned	3
fetched	3
reused	0
failed	0
named	3
candidates	3
mentions	3
capped	0

Rank	Rival	Seeds	Mentions	First seed	Kind	Link
1	Claude Code	1	1	col0-claude	named	https://code.claude.com/docs/en/overview
2	Codex CLI	1	1	col1-codex	named	https://raw.githubusercontent.com/openai/codex/main/README.md
3	Live-SWE-agent	1	1	col3-live	named	https://arxiv.org/abs/2511.13646

Exclusive and shared capabilities

One capability is graded on one board; every other board references it. The fleet duplicate census (nx_rivalcensus dupes, on the fleetdupes clock row) reads every board's matrix and names each capability graded as a row on two or more boards, with a home derived from evidence: rung when exactly one board's plan contracts the symbol (that board is starred), gates when no plan contracts it and exactly one board's executed gate map runs the organ's gate (that board is plussed), ambiguous when several boards carry the deciding evidence (a decision for their seats), and none when no evidence decides it, with the reason beside it: unmapped (no board of the duplicate declares a gates map; declare the maps) or unproven (a map exists and runs none of the organ's gates; gate the organ on its proving board, or declare it a shared library by policy). A home elsewhere means this board should carry an @xrow reference to the home board's row rather than a graded row of its own; a home here means the other boards should. The table is this board's slice of one estate-wide census: the shape is shared, the rows are the board's.

This board grades 41 capabilities: 23 exclusive to it and 18 shared with other boards. Shared rows on this page: 18 = home here 1 + home elsewhere 6 + ambiguous 6 + undecided 5 + refused 0 (partition 18); undecided = unmapped 0 + unproven 5 + unevidenced 0; homonyms 0; since the previous census: new 0, resolved 0. Census FRESH age_s=6971.

Capability	Organ	Boards grading it	Home	Rule	Evidence	Standing on this board
fc_emit	`nx_forge_ctx.nx`	coding, extllm, autodev	—	none	unproven	undecided
fg_run	`nx_forge.nx`	coding*, extllm, autodev	coding	rung	plan	home elsewhere adopt: `@xrow coding\|fg_run\|home by rung (fleet duplicate census)`
load_named_q16	`nx_nofloat_llm.nx`	coding*, autodev	coding	rung	plan	home elsewhere adopt: `@xrow coding\|load_named_q16\|home by rung (fleet duplicate census)`
fg_best_of_n	`nx_forge.nx`	coding, extllm, autodev, engineshift	—	ambiguous	plan	ambiguous
ci_query_top1	`nx_code_index.nx`	coding*, autodev	coding	rung	plan	home elsewhere adopt: `@xrow coding\|ci_query_top1\|home by rung (fleet duplicate census)`
fg_multi_file	`nx_forge.nx`	coding, autodev	—	ambiguous	plan	ambiguous
fg_polyglot_bench	`nx_forge.nx`	coding, autodev	—	ambiguous	plan	ambiguous
db_open_count	`nx_debt.nx`	pmdash, autodev	—	none	unproven	undecided
cd_show_lost	`nx_contentdiff.nx`	devmgmt, codequality, autodev, capregistry, codeeffectiveness	—	none	unproven	undecided
gb_try	`nx_gate_bite.nx`	devmgmt+, codequality, autodev*, codeeffectiveness	autodev	rung	plan	home here
os_jrnl	`nx_organ_ship.nx`	devmgmt+, autodev	devmgmt	gates	gates	home elsewhere adopt: `@xrow devmgmt\|os_jrnl\|home by gates (fleet duplicate census)`
fc_alert_score	`nx_failclass.nx`	failmodes*, autodev, codeeffectiveness	failmodes	rung	plan	home elsewhere adopt: `@xrow failmodes\|fc_alert_score\|home by rung (fleet duplicate census)`
sg_verdicts	`nx_swebv_grade.nx`	extllm, autodev	—	none	unproven	undecided
xl_call	`nx_extllm_call.nx`	extllm, autodev, engineshift	—	ambiguous	plan	ambiguous
xl_egress_class	`nx_extllm.nx`	extllm*, autodev, engineshift	extllm	rung	plan	home elsewhere adopt: `@xrow extllm\|xl_egress_class\|home by rung (fleet duplicate census)`
xl_distill_corpus	`nx_extllm.nx`	extllm, autodev, engineshift*	—	ambiguous	plan	ambiguous
lf_gen_t	`nx_autofix_auto.nx`	autodev, engineshift	—	none	unproven	undecided
af_sampler_tier	`nx_autofix_auto.nx`	autodev, engineshift	—	ambiguous	plan	ambiguous

Gaps from the record — what the estate does that no board carries

The record census (nx_goalmap record) reads the invoked-tool population and every plan queue row and files each organ or directive that NO matrix, plan or gates row names. A row here is a callout the boards missed: adjudicate it onto a board or declare it infrastructure. Census state STALE (age 1396076 s), sources read 4 of 7 declared — a BLIND census is a FLOOR: unread sources can only add rows.

kind	name	board	source	evidence

rows shown 0|this board's directives 0|estate-wide un-boarded organs 452 (listed in full on /compare/ecosystem)|census rows 639|malformed 0 (counted, never rendered)

References

Beyond a link list. Every reference below resolves twice — the publisher's copy and, where banked, the estate's own non-rottable library mirror with a content pin — and carries its evidence class plus the exact claim on this page it grounds. Keyed marks like [key] in the matrix notes jump here. A dash means honestly absent, never assumed.

[harnessthesis] Zhang, Wang, Ge, Xu, Hamm and Reddy. Stop Comparing LLM Agents Without Disclosing the Harness (position paper, submitted 2026-05-07). The Binding Constraint Thesis: for long-horizon tasks across models of comparable frontier capability, harness configuration governs more performance variance than model choice, with measured cases of model ranking reversal; proposes a harness disclosure standard and a variance decomposition protocol. arXiv 2605.23950. publisher · read in our library knowledge/fetched/cmp_autodev_harnessthesis.html · pin he6f975369dc5da48550dbf83894d8baad17eedf57200a9ee5481094df989aad3 · accessed 2026-08-27 · published-paperGrounds: RIGOR: harness disclosure manifest hashed beside every result
[metaharness] Lee, Nair, Zhang, Lee, Khattab and Finn. Meta-Harness: End-to-End Optimization of Model Harnesses (submitted 2026-03-30). The harness is the code that determines what information to store, retrieve and present to the model; an agentic proposer with filesystem access to prior candidates automates harness optimisation and surpasses hand-engineered baselines on TerminalBench-2. arXiv 2603.28052. publisher · read in our library knowledge/fetched/cmp_autodev_metaharness.html · pin he25f14e1e507216df47702e5410dc3d208b8e0d0d3dc4a875c2a2911f7d280c1 · accessed 2026-08-27 · published-paperGrounds: EVOLVE: harness evolution as proposals scored on a held-out split
[ahe] Lin et al. Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses (submitted 2026-04-28, revised 2026-05-18). Closed-loop harness evolution on three observability pillars; Terminal-Bench 2 pass at 1 lifted from 69.7 to 77.0 over ten iterations, and on SWE-bench Verified top performance with 12 percent fewer tokens. arXiv 2604.25850. publisher · read in our library knowledge/fetched/cmp_autodev_ahe.html · pin h62fcbc1aebd90defa93c346fa1e0d3b518e569c20845e3116a945cb800c53d3a · accessed 2026-08-27 · published-paperGrounds: EVOLVE: harness evolution as proposals scored on a held-out split
[harnessevalrethink] Wang et al. Rethinking the Evaluation of Harness Evolution for Agents (submitted 2026-07-14). Two flaws named in the field: harness search is not compared against matched-budget simpler baselines, and it is evaluated on the same benchmark it searched on; measured on Terminal-Bench 2.1 with GPT-5.4 and Claude Opus 4.6, automatic harness evolution does not consistently outperform simple test-time scaling and generalises poorly to held-out tasks. arXiv 2607.12227. publisher · read in our library knowledge/fetched/cmp_autodev_harnessevalrethink.html · pin h85f0b5cf005a157cb4f7b866ff58d7a445fe22cc73dffb738fb6b5ddde3660ac · accessed 2026-08-27 · published-paperGrounds: EVOLVE: the matched-budget and held-out accept rule this board pre-declares
[liveswe] Xia, Wang, Yang, Wei and Zhang. Live-SWE-agent: Can Software Engineering Agents Self-Evolve on the Fly? (submitted 2025-11-17). An agent that starts with bash tools and modifies its own scaffold implementation while solving problems; 77.4 percent on SWE-bench Verified without test-time scaling and 45.8 percent on SWE-Bench Pro. arXiv 2511.13646. publisher · read in our library knowledge/fetched/cmp_autodev_liveswe.html · pin h478579f688f859d81f4022a55b971b5846aa6992631ca0055254b09647fc4af3 · accessed 2026-08-27 · published-paperGrounds: The Live-SWE-agent column and the self-evolving-scaffold rung
[lilharness] Weng. Harness Engineering for Self-Improving Coding Agents (blog essay, 2026-07-04). A harness is the system surrounding a base model that orchestrates execution and decides how the model thinks and plans; near-term recursive self-improvement proceeds by evolving the harness layer through automated search rather than by rewriting weights; names Meta-Harness, Self-Harness, AHE, AlphaEvolve and ADAS as the systems doing it. publisher · read in our library knowledge/fetched/cmp_autodev_lilharness.html · pin h45bc3793c36e42f79749f84af510761798b85960ad18d3123d86f3a403e247ac · accessed 2026-08-27 · published-courseGrounds: CONTEXT: the harness component list this board measures itself against
[ttcagentic] Kim et al. Scaling Test-Time Compute for Agentic Coding (submitted 2026-04-16). Rollouts converted to structured summaries so that Recursive Tournament Voting (parallel) and Parallel-Distill-Refine (sequential) apply to long-horizon agents; Claude-4.5-Opus from 70.9 to 77.6 percent on SWE-Bench Verified and Terminus 1 from 46.9 to 59.1 on Terminal-Bench v2.0. arXiv 2604.16529. publisher · read in our library knowledge/fetched/cmp_autodev_ttcagentic.html · pin h6ceef45a19fcda26c8bb90d55d365427ab65a991a1eb02d17f30466e5b221316 · accessed 2026-08-27 · published-paperGrounds: PROPOSE: sampling and selection against the verifier
[runornot] Lin et al. To Run or Not to Run: Analyzing the Cost-Effectiveness of Code Execution in LLM-Based Program Repair (submitted 2026-06-25). 7,745 SWE-bench leaderboard traces and 3,000 end-to-end attempts over 200 instances with Claude Code, Codex and OpenCode under four execution paradigms: agents average 8.8 test runs per task ranging 2 to 19, prohibiting execution narrows the resolve-rate gap by only 1.25 points while saving tokens and time, and the benefit concentrates on a subset of instances. arXiv 2606.26978. publisher · read in our library knowledge/fetched/cmp_autodev_runornot.html · pin hf05b212a46e782b0c126b8c9c40cafc28d95a124330956e48a6b2c51f7012ff4 · accessed 2026-08-27 · published-paperGrounds: VERIFY: the execution budget measured per task class
[adi] Xiang, Xu, Chu, Tian and Zhang. Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis (submitted 2026-04-27, FSE 2026). The Agent-centric Debugging Interface with a function-level Frame Lifetime Trace resolves 63.8 percent of SWE-bench Verified at 1.28 dollars per task on Claude-Sonnet-3.7 and lifts existing agents by 6.2 to 18.5 percent. arXiv 2604.24212. publisher · read in our library knowledge/fetched/cmp_autodev_adi.html · pin h1d19de173a1db32df5a5a4be610701e886f991b1b70c34a7648d1695a98af07d · accessed 2026-08-27 · published-paperGrounds: LOCALIZE: function-level dynamic evidence beside static retrieval
[brtcogen] Cheng et al. Dynamic Cogeneration of Bug Reproduction Test in Agentic Program Repair (Google, submitted 2026-01-27, revised 2026-08-21). On 120 human-reported bugs, instructing the repair agent to emit the bug-reproduction test in the same patch as the fix yields reproduction tests for at least as many bugs as a dedicated test agent without lowering the plausible-fix rate, and patch selectors that account for test change select better patches. arXiv 2601.19066. publisher · read in our library knowledge/fetched/cmp_autodev_brtcogen.html · pin h635af7312b522354983d86304fa02c6ae30b53426753bb8bc460ce63ec5fb400 · accessed 2026-08-27 · published-paperGrounds: REPRO: the reproduction tooth cogenerated with every fix
[chainswe] Jin et al. ChainSWE: Benchmarking Coding Agents on Multi-Bug Software Maintenance (submitted 2026-07-01). 304 chronologically chained issues across 54 Python projects mined from six SWE-bench-family datasets; agents evaluated on sequential dependent fixes in one shared codebase drop by up to 70 percent as the chain lengthens. arXiv 2607.02606. publisher · read in our library knowledge/fetched/cmp_autodev_chainswe.html · pin hfcca15585b04cae26be323f5b4c90713094a4cc53f69cfd89edf74c67ded85d0 · accessed 2026-08-27 · published-paperGrounds: CHAIN: dependent tasks without a reset and the per-chain-length number
[swerebenchv2] Badertdinov, Nekrashevich, Shevtsov and Golubev. SWE-rebench V2: Language-Agnostic SWE Task Collection at Scale (submitted 2026-02-27, revised 2026-06-01). An automated language-agnostic pipeline yielding 32,079 executable tasks across 20 languages from 3,617 repositories with pre-built images, unsound instances filtered by an ensemble of LLM judges, plus 120,000 further tasks with metadata; built to relieve the task scarcity that constrains RL training of agents. arXiv 2602.23866. publisher · read in our library knowledge/fetched/cmp_autodev_swerebenchv2.html · pin h8bbd76e19cdd1626096a22d67ee80eb0572ab67c894d489cbd54e6c30cb39106 · accessed 2026-08-27 · published-paperGrounds: REFEREE: the post-cutoff multilingual task source
[swebenchpro] Deng et al. SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks? (submitted 2025-09-21, revised 2025-11-14). 1,865 problems from 41 actively maintained repositories split into public, held-out and commercial sets; long-horizon tasks that take a professional hours to days and patch multiple files; human-verified and presented as a contamination-resistant testbed. arXiv 2509.16941. publisher · read in our library knowledge/fetched/cmp_autodev_swebenchpro.html · pin h21f69a25e615c4223357fe653663aeb94f962732cbbb1277e827bd3014f9df53 · accessed 2026-08-27 · published-paperGrounds: REFEREE: the held-out long-horizon bar and the multi-file rung
[hal] Kapoor et al. Holistic Agent Leaderboard: The Missing Infrastructure for AI Agent Evaluation (submitted 2025-10-13). A standardised harness running 21,730 rollouts across 9 models and 9 benchmarks for about 40,000 dollars; higher reasoning effort reduced accuracy in the majority of runs; LLM-aided log inspection found agents searching for the benchmark on HuggingFace instead of solving the task; all 2.5B tokens of logs released. arXiv 2510.11977. publisher · read in our library knowledge/fetched/cmp_autodev_hal.html · pin hd1184a420b01d05e12edfc73b31048c0957348296eb5cf6c34b420267ababcfb · accessed 2026-08-27 · published-paperGrounds: LEDGER and GUARD: logs as the audit and benchmark-gaming as a named failure
[dontpassk] Samandar et al. Don't Pass at k: A Bayesian Framework for Large Language Model Evaluation (arXiv 2510.04265, under review ICLR 2026). Replaces pass at k and avg at N with posterior success probabilities and credible intervals under a Dirichlet prior, closed-form for any weighted rubric; declares a difference only when intervals do not overlap and allocates further samples until intervals reach a pre-specified width; faster convergence and more stable rankings than pass at k on synthetic and math benchmarks. publisher · read in our library knowledge/fetched/cmp_autodev_dontpassk.html · pin h8cf03e8def93f54ba6c87e717816361daa40eb863a8379b2013688dd3a286ced · accessed 2026-08-27 · published-paperGrounds: RIGOR: the interval and the decision rule every published number carries
[statprecipice] D'Oro et al. Computer Use at the Edge of the Statistical Precipice (submitted 2026-05-07). A 1 MB replay script that never observes the screen outperforms frontier models on static benchmarks, with expected success exactly the source agent's pass at k in deterministic environments; the PRISM principles (privileged verification, realistic environments, integrity-checked configurations, sandboxed execution, multifactorial variability) and Wilson score intervals with hierarchical bootstrap for the nested structure of agent benchmarks. arXiv 2605.08261. publisher · read in our library knowledge/fetched/cmp_autodev_statprecipice.html · pin hf1c4c6ed1d17b82a952ebcd3af187473c0b7afc840e861159499f1145e242e19 · accessed 2026-08-27 · published-paperGrounds: RIGOR and CONTROL: the replay null control and the hierarchical interval
[swebenchcase] Martinez and Franch. What's in a Benchmark? The Case of SWE-Bench in Automated Program Repair (submitted 2026-02-04). First study of the Lite (79 entries) and Verified (133 entries) leaderboards: most entries come from industry, proprietary LLMs dominate with the Claude family at the top, and academic open-source entries remain competitive. arXiv 2602.04449. publisher · read in our library knowledge/fetched/cmp_autodev_swebenchcase.html · pin h45932060b651ab7ae1db2288e4da062979b7be244394bbe92b525bc963958218 · accessed 2026-08-27 · published-paperGrounds: REFEREE: why Verified is a sanity check and not the frontier for this board
[swezerohero] Ludwig, Ahmad, Majumdar and Ginsburg. From SWE-ZERO to SWE-HERO: Execution-free to Execution-based Fine-tuning for Software Engineering Agents (submitted 2026-04-02, revised 2026-05-06). Two-stage SFT over 300,000 execution-free and 13,000 execution-based trajectories distilled from open-weight frontier models; SWE-HERO-32B resolves 62.2 percent of SWE-bench Verified and 44.1 percent of SWE-bench Multilingual zero-shot from Python-only training. arXiv 2604.01496. publisher · read in our library knowledge/fetched/cmp_autodev_swezerohero.html · pin h2fe810a07d9da7fbc919617f7dfeb5979b17f5a1d140baa999b561712c1ba926 · accessed 2026-08-27 · published-paperGrounds: DISTILL: what a verified-trajectory corpus is worth and where the terms let us build one
[sweagent] Yang, Jimenez, Wettig, Lieret, Yao, Narasimhan and Press. SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering (submitted 2024-05-06, final 2024-11-11). A custom agent-computer interface for creating and editing files, navigating repositories and executing tests; pass at 1 of 12.5 percent on SWE-bench and 87.7 percent on HumanEvalFix at the time. arXiv 2405.15793. publisher · read in our library knowledge/fetched/cmp_autodev_sweagent.html · pin h1589f706b0380d637196a65970b23cb89037465fe6d883af1007e34a41d8bdbf · accessed 2026-08-27 · published-paperGrounds: LOCALIZE and VERIFY: the agent-computer-interface lineage every rival column shares
[openhands] Wang et al. OpenHands: An Open Platform for AI Software Developers as Generalist Agents (submitted 2024-07-23, revised 2025-04-18, ICLR 2025, MIT licence). Agents that write code, use a command line and browse; sandboxed environments for code execution; multi-agent coordination; benchmarks incorporated in the platform and evaluated over 15 tasks including SWE-bench and WebArena. arXiv 2407.16741. publisher · read in our library knowledge/fetched/cmp_autodev_openhands.html · pin h2b9f10c47d1964713d513d8652d75cc988bbd3d7415f89abd8446bfab6e4e3c5 · accessed 2026-08-27 · published-paperGrounds: The OpenHands column: sandbox, tools and evaluation harness
[claudecode] Anthropic. Claude Code documentation, Overview (fetched 2026-08-27). An agentic coding tool that reads the codebase, edits files, runs commands and integrates with development tools across terminal, IDE, desktop and web; git staging, commits and pull requests; MCP connections; CLAUDE.md instructions, auto memory, skills and hooks; parallel subagents and an Agent SDK; headless piping in CI; scheduled routines. publisher · read in our library knowledge/fetched/cmp_autodev_claudecode.html · pin h48897aa2d9b166167b54b333c2c440a6e6c58daf526f9e83363cbef66c6a18c6 · accessed 2026-08-27 · vendor-docGrounds: The Claude Code column: the product surface as the vendor describes it
[codex] OpenAI. Codex repository README (openai/codex, main, fetched 2026-08-27). Codex CLI is a coding agent from OpenAI that runs locally on your computer; installation for Mac, Linux and Windows, editor integrations, sign-in with ChatGPT plans or an API key. The README as mirrored does not describe sandboxing, approval modes, cloud tasks or MCP, so those cells are coded from silence and say so. publisher · read in our library knowledge/fetched/cmp_autodev_codex.md · pin hba4e1f69ff48386e72a9c5e1edaf76aad64a475c2d51af79ccba6d1128261ba7 · accessed 2026-08-27 · vendor-docGrounds: The Codex column: a thin mirror, coded honestly as such
[dgm] Zhang, Hu, Lu, Lange and Clune. Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents (submitted 2025-05-29, revised 2026-03-12). A self-improving system that modifies its own code and validates each change empirically on coding benchmarks from an archive of agents; SWE-bench from 20.0 to 50.0 percent and Polyglot from 14.2 to 30.7; all experiments under sandboxing and human oversight. arXiv 2505.22954. publisher · read in our library knowledge/fetched/cmp_autodev_dgm.html · pin h30ff41b60abf5df7cdd7af79d2b22eb92a2777afedb85dd297874327961f4649 · accessed 2026-08-27 · published-paperGrounds: EVOLVE and SANDBOX: empirical validation of self-modification under isolation
[alphaevolve] Novikov et al. AlphaEvolve: A coding agent for scientific and algorithmic discovery (DeepMind white paper, submitted 2025-06-16). An evolutionary coding agent that continuously receives feedback from one or more evaluators and iteratively improves an algorithm; a 4x4 complex matrix multiplication in 48 scalar multiplications, the first improvement over Strassen in 56 years, and deployed data-centre scheduling and accelerator-design results. arXiv 2506.13131. publisher · read in our library knowledge/fetched/cmp_autodev_alphaevolve.html · pin h7a0d9acc7002b1960fea119a059a925a6c20d27edd8233777aa0179cb5e8f451 · accessed 2026-08-27 · published-paperGrounds: EVOLVE: the evaluator-driven loop our gate-permil fitness already mirrors
[codeharness] Ning et al. Code as Agent Harness (survey, 42 authors, submitted 2026-05-18). Code as the operational substrate for agent reasoning, acting, environment modelling and execution-based verification; open challenges named include evaluation methodology, verification with incomplete feedback, regression prevention, shared state consistency and human oversight. arXiv 2605.18747. publisher · read in our library knowledge/fetched/cmp_autodev_codeharness.html · pin h21feb4689d1535ec99d95edf34e5b363fb68165b7572f9bb45761f89687ad66b · accessed 2026-08-27 · published-paperGrounds: VERIFY: regression prevention as a named open challenge the promote ruler answers
[swebench] Jimenez et al. SWE-bench: Can Language Models Resolve Real-World GitHub Issues? (arXiv 2310.06770). 2,294 real GitHub-issue problems across Python repositories; a patch is RESOLVED only when the FAIL_TO_PASS tests pass and the PASS_TO_PASS tests still pass, with the test suite as the sole judge. publisher · read in our library knowledge/fetched/cmp_autodev_swebench.html · pin hcd170cab3a56f891ee94070d5748d923685f2a84a91201d6336a221436fd0f9f · accessed 2026-08-27 · published-paperGrounds: VERIFY: the contract our sovereign analog adopts verbatim
[oxalpha] OpenRouter. Ox Alpha model page (stealth/ox-alpha, released 2026-08-20; mirror pinned by the extllm domain 2026-08-24). A reasoning model designed for coding, sustained agentic work and production workloads; 1,048,576-token context; the page banner states prompts and completions are retained by the provider and not used for training and defers all other use to the Stealth Model Terms. publisher · read in our library knowledge/fetched/cmp_extllm_oxalpha.html · pin h108b8c85dfed1a9b5726af14f6ca955c1a7f5d56c3d6161f70009d88a42586e2 · accessed 2026-08-24 · vendor-docGrounds: SAMPLER: the free frontier sampler behind the warden, capability facts only
[stealtheula] OpenRouter. Stealth Program End User License Agreement (updated 2026-07-06; mirror pinned by the extllm domain 2026-08-24). Stealth models are offered free of charge for purposes of collecting User Content for use in Stealth Model training and improvement, with an irrevocable perpetual licence to the provider, and Exhibit A forbids publicly disseminating confidential technical information regarding the performance of the models. publisher · read in our library knowledge/fetched/cmp_extllm_stealtheula.html · pin hfed941390d8fbcb7b06d92337a4b367997e6de39452edec152dfee899f2f6a31 · accessed 2026-08-24 · vendor-docGrounds: GUARD: why no ox-alpha performance number appears on this page and why the warden classifies every prompt byte

generated by nx_swcompare_matrix (sovereign NishiLang organ) from knowledge/compare/autodev.matrix · source checks show implementation presence; runtime and user-outcome evidence are reported separately · JavaScript supports page controls