Nishi Family › Compare › Deep Research and Answer Engines

Nishi Compare · capabilities, evidence and delivery

Deep Research and Answer Engines

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

Nishi vs Perplexity and DeepResearcher and Tongyi DeepResearch and deep-searcher

Overview

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

Where we are. The sovereign research engine is real and deterministic: fetch-and-bank over own TLS, BM25 and dense and PPMI retrieval, a pretrained sovereign Qwen reader measured on-protocol (SQuAD 510, Hotpot 505 vs mechanical 250), F1 scoring, a compounding findings ledger, corpus semantic retrieval, and adversarial refutation that is AUTO-INVOKED in-run: every numeric claim graded against its cited source and cross-validated across independent origins before publish. Two exceeds: fully sovereign at zero dollars per query, and integer-deterministic crash-resumable with no link-rot. Behind the field on: RL-trained agency, fluent synthesis, a trained cross-encoder reranker, multimodal output, and a served product API.

Where we need to go. A research engine someone else can call: a served API over the existing pipeline, fluent synthesis written by the sovereign model but GATED by the same refutation that already guards numbers, a trained reranker that beats the lexical arm on a pre-declared ruler, charts and figures in the report, and an agent whose policy is learned from its own verified runs -- never a hosted LLM in the trust path.

The unit. 1 u = one measured session-leg (estate calibration: graphics R21 in one leg 2026-08-15). Local evidence: auto-invoked refutation and multi-source cross-validation each landed gated (4/4, 5/5, 7/7, 10/10) in one leg apiece on 2026-07-14; estimates are relative to those.

Where we are: 5 open rungs. Retrieval, reading and verification present and honest; product surface and learned components absent. Counts measured at emit below this line.

Cost to a callable product: 1.5 u. DR1 the served research API over nx_research_unified -- the pipeline already runs end to end, the door is missing.

Cost to fluent, gated synthesis: 2 u. DR2 the sovereign model writes the answer; nxr_assess grades every claim in it before it ships, exactly as it grades extractive output today.

Cost to learned quality: 8 u. DR3 the trained cross-encoder reranker (accept rule pre-declared against BM25 321 permil), DR4 charts in the report, DR5 the learned agent policy over the verified-run ledger.

Research bar. Perplexity is measured on served API, polished synthesis, multimodal answers. Theirs: the product bar. Ours: DR1 DR2 DR4 measured on this page.

Research bar. DeepResearcher (arXiv 2504.03160) is measured on GRPO-trained research agent, multi-source cross-validation. Theirs: the learned-agent bar. Ours: DR5; cross-validation already matched.

Research bar. arXiv query API is measured on whether a machine can ask the scholarly frontier a question without a key or a browser. Theirs: a public Atom API with a Boolean field grammar. Ours: DN0 already composes it as a data-declared source row [arxiv-api].

Research bar. OpenAlex works API is measured on a second key-free index, which is what makes a wide sweep a pace problem rather than a quota problem. Theirs: free REST over the open catalogue, no key. Ours: DN1 must admit it before the sweep can run fleet-wide [openalex-api].

Latest recorded release

No valid dated release entry is recorded for this domain.

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

16 of 25 capability source symbols found|4 of them authored stronger-claims|9 open|coverage 640/1000|adoption 7 full / 9 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 = 16 vs present 16 · named + unnamed + withheld = 9 vs open 9 · both reconcile

liar-kill conj=GPQN · all four conjuncts held

graded document: BUILDROOT tree, 19364 bytes · gates map: PRIMARY · stamped 2d 13h ago · source ../knowledge/status/evstamp_deepresearch.verdict

Gap class	What it is, and the work it names
NONE	No gap class fires on this board: every published claim is grounded, the declared gates ran and were green, no watch contract is unnamed or already landed, and the stamp is inside its TTL. That is a statement about evidence, not about depth, scale or polish.

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 deepresearch 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_INCONSISTENT. 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_INCONSISTENT. 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.
Unified refutation in-run (DB0)	nxr_assess	Every numeric claim graded vs its source and cross-validated before publish -- LANDED gated	Organ	0 u
Sovereign reader (DB1)	qwx_extract	On-protocol SQuAD 510, Hotpot 505 -- LANDED	Organ	0 u
Lexical retrieval (DB2)	idf_q10	BM25 baseline, integer-deterministic -- LANDED	Organ	0 u
Served research API (DR1) after DB0	rapi_serve	An authenticated HTTPS door that runs the unified pipeline for a question and returns the graded report with per-claim verdicts; gate proves a request over the edge returns a report whose claims carry their sources and a fabricated-number fixture is refused by the same path	Organ	1.5 u
Gated synthesis (DR2) after DB0,DB1	ls_synthesize	The sovereign model composes prose from the extractive findings; every sentence with a number or a citation is graded by nxr_assess and a discredited claim forces rewrite or STOP-INCOMPLETE, never ships; gate proves a planted false number in the synthesis is caught and the report abstains	Organ	2 u
Trained cross-encoder reranker (DR3) after DB2	nr_cross_score	A no-float cross-encoder scoring (query, passage) pairs, trained on the ledger's verified pairs; ACCEPT RULE declared before the run: must EXCEED the BM25 arm (321 permil on the BEIR ruler) on the same queries or it stays UNWIRED, as the semppmi and RRF arms did	Organ	3 u
Charts in the report (DR4) after DR2	mr_chart_emit	Numeric findings rendered as sovereign SVG charts inside the report, each chart citing the claims it plots; gate proves a chart's plotted values equal the graded claims byte-for-byte and a chart of an ungraded number is refused	Organ	1.5 u
Learned agent policy (DR5) after DB0,DR3	ra_policy_step	A policy that chooses the next action (fetch, read, refute, stop) trained on the verified-run ledger through the no-float autograd; accept rule: higher on-protocol F1 per fetch than the hand-written loop on a held-out question set, or UNWIRED	Organ	3.5 u

Milestones

Milestone	Rungs	Cumulative
M1 · Callable	DR1	1.5 u
M2 · Fluent and gated	DR2,DR4	5 u
M3 · Learned	DR3,DR5	11.5 u

Rung	Closes with	Definition of done (pre-declared)	Executor	Est.
The noticer itself (DN0)	fs_deficit	LANDED 2026-08-20. Seeds derived from the domain's own board, outside pulled through the sovereign fetcher, findings diffed against our rows and refs, GAP PROPOSALS filed to a worklist and the frontierprop- plane. Done-rule: gate GREEN with a negative control proving a no-gap board issues ZERO fetches and ZERO proposals, and a tooth proving a run never writes the board it reads	Organ	0 u
Fleet sweep on a beat (DN1) after DN0	fs_sweep_all	Every /compare domain scanned on a cadence instead of one. Done-rule: a per-source pace budget that the run PRINTS and obeys, a second key-free index admitted as a source row, and a full-fleet run whose outbound request count matches rows times sources times domains exactly -- a sweep that cannot state its own request count is a load incident waiting to happen	Organ	1.5 u
Proposal adjudication in the hive (DN2) after DN0	fs_adjudicate	A seat accepts or rejects a proposal by ONE call that closes the plane row and, on accept, emits the ready-to-paste ref row with the mirror already fetched and pinned. Done-rule: an accepted proposal produces a refs row that nx_compare_refs_gate passes on first run, and a rejected one leaves the board byte-identical	Organ	1 u
Mirror bound to its own URL (DN3)	crg_mirror_url_bound	The referee proves a mirror is a capture of the url in ITS OWN ROW, not merely that some file exists and its pin matches. Done-rule: the four 2026-08-18 charsim rows that pass today must FAIL this tooth, and every honestly-fetched row must still pass -- the bite is already sitting in the corpus	Organ	1 u

Milestones

Milestone	Rungs	Cumulative
M4 · The estate notices its own gaps	DN0,DN1,DN2,DN3,DN4	5.0 u

Rung	Closes with	Definition of done (pre-declared)	Executor	Est.
Drift checked across every register on a beat, with a verdict something can read (DN4) after DN3	rd_beat	The four-state organ becomes the one that RUNS, and its answer survives the run. Done-rule, all five and each independently checkable: the PROMOTED artifact prints a dynamic field in its envelope, so the state that exists in source is the state being served and the source is no longer ahead of the binary; the run truncate-writes a verdict log with a canonical last line, because a census whose verdict is never persisted has no artifact to disagree with and nothing on a beat or in a deploy-ready row can consume it; a rostered nx_refdrift_gate exists and is BITE-PROVEN in both directions, since a comparator that answered same for everything would report a permanently clean fleet and the self-verb selftest has never been mutated; the sweep covers every refs register across the fleet under a per-source pace budget the run PRINTS and OBEYS, because outbound cost is rows times registers and a sweep that cannot state its own request count is a load incident waiting to happen; and DRIFTED is RATCHETED and self-baselined on first sight rather than armed, because arming it fleet-wide would turn every domain holding an aged mirror RED in one step and teach everyone to ignore it, while DYNAMIC is NEVER counted as a defect and every DYNAMIC url is NAMED, since a count without a worklist is not actionable	Organ	1.5 u

Ladder to SOTA

Two dated targets, and every rung's role toward them. Best in class is the proven, deployed leader today; the frontier is the research edge as of its month. Each target is dated by a sotabar row and judged against the current month by the bar-age ruler (FRESH, ATTESTED with the fallback month named, or refused). Each rung is substrate (what the target's own methods consume), an arm (a sovereign arm that triangulates without competing for the number), a contender (its done-rule is a number against the target) or superseded (kept as capability, retired as a claim), and carries the rungs it stands on.

best-in-class (DRT1), dated by DRB1 = 2026-09 UNATTESTED. A research engine someone else can call: a served API over the existing sovereign pipeline, returning a graded report whose every claim carries its source, with fluent synthesis and figures -- matching the product surface Perplexity ships, with none of the hosted-LLM dependency in the trust path. ref arxiv_dr_202609

frontier (DRT2), dated by DRB2 = 2026-09 UNATTESTED. An agent that learns its own procedure from its verified-run ledger, calibrates its confidence, and cannot write into the report or the memory until the critic has graded what it wrote -- so that faithfulness is a property of the mechanism rather than of the model's mood. OUR HEAD START IS THE CRITIC, WHICH THE FIELD IS STILL SPECIFYING; OUR GAP IS THAT THE CRITIC AUDITS NUMBERS AND THE FIELD IS AUDITING THE WHOLE REPORT. ref arxiv_dr_202609

Rung	Role	Toward	Stands on	Why
DB0	substrate	DRT2	-	The auto-invoked refutation is what every faithfulness claim on this board stands on; without it a fluent report is unfalsifiable.
DB1	substrate	DRT1	-	The sovereign reader is the extraction layer the served product returns answers from.
DB2	substrate	DRT1	-	Deterministic lexical retrieval is the recall floor every later arm must beat rather than replace.
DR1	contender	DRT1	DB0	Its done-rule IS the target's number: a callable door returning a graded report is what best-in-class means here.
DR2	contender	DRT2	DB0,DB1	Gated synthesis is the faithfulness bar directly -- the writing stage becomes the audited object, which is what the 2026-09 bar describes.
DR3	arm	DRT1	DB2	A trained reranker triangulates quality against the lexical arm and is accept-rule gated, so it cannot flatter the product number by replacing the measurement.
DR4	contender	DRT1	DR2	Figures in the report are part of the product surface the best-in-class rival ships.
DR5	contender	DRT2	DB0,DR3	A learned policy over the verified-run ledger is exactly the procedural-experience direction the month names.
DN0	substrate	DRT2	-	The noticer is what lets the estate see a frontier move without being told, which is the precondition for any target staying dated.
DN1	contender	DRT2	DN0	A fleet sweep is the difference between noticing one domain's gaps and noticing the estate's. ITS TWO BLOCKING CONDITIONS ARE NOW SUPPLIED BY /compare/researchreach: openalex is admitted as a key-free source row and RR3 is the per-source pace budget, so this rung and researchreach RR6 are ONE capability carried as two contracts and should be reconciled when either lands.
DN2	arm	DRT2	DN0	Adjudication turns a proposal into a board row without a seat re-doing the research; it serves the target without competing for its number.
DN3	substrate	DRT2	-	A mirror bound to its own url is the substrate under every citation this board makes; an unbound mirror makes faithfulness unmeasurable.
DN4	substrate	DRT2	DN3	Drift on a beat is what keeps a once-true citation from becoming a silent lie, which is the time axis of the same faithfulness target.

Rung	Closes with	Definition of done (pre-declared)	Executor	Est.
Research output preserves qualitative findings and reproducible quantitative claims (DR6) after DR4,researchreach:RR15	_ABSENT_:research_output_evidence_accept	SOURCE-BOUND PLANNING CONTRACT 2026-09-24. Compose the existing graded-claim report/chart and paper-evidence owners. A report binds question, method, input versions, qualitative observations, codebook version, numeric derivations, uncertainties, failures and reviewer decisions; no number replaces its original narrative. Demonstrate a joint display where each plotted value resolves to the source observation and derivation, and an invented value, mismatched codebook or omitted failed run is rejected or shown unresolved. Reproduce the declared result from retained inputs and distinguish independent reviewer agreement from truth or delivery acceptance. No universal photoreal percentage, invented threshold, model training or automatic paper publication. Existing documentation/Compare surfaces carry the dated result and unfinished work; effort unestimated.	Workflow	unestimated u

Rung	Role	Toward	Stands on	Why
DR6	substrate	DRT2	DR4,researchreach:RR15 UNRESOLVED(no-matrix-row)	Reproducible qualitative and quantitative evidence is required before a research faithfulness claim can be independently reviewed.

Ladder verdict. LADDERED. Targets, rung roles and bars are counted in the tables below, each partition re-derived beside its total. Bars read STALE against the current month 2026-10.

Ladder verdict, counted (17 rows, more than the 12-row disclosure bar)

Ladder verdict, counted
Class	Count	Note
targets	2	sotatarget rows on the plan
best in class	1	targets whose class is best-in-class, by code
frontier	1	targets whose class is frontier, by code
undated	0	targets naming no bar that exists
bad class	0	targets whose class is neither word
duplicate	0	target ids an earlier row already carries
rungs	14	rung rows on the plan
placed	14	rungs carrying a rungrole row
unplaced	0	rungs with no role toward any target
bad role	0	rungrole rows naming a role the ladder does not know
bad target	0	rungrole rows naming a target no sotatarget row declares
rungs, summed	14	RECONCILED: the parts sum to the total 14
substrate	7	what the target's own methods consume
arm	2	a sovereign arm that triangulates without competing for the number
contender	5	its done-rule is a number against the target
superseded	0	kept as capability, retired as a claim
placed, summed	14	RECONCILED: the parts sum to the total 14

Bar freshness, counted
Class	Count	Note
bars	2	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	2	nobody looked this month
malformed	0	a sotabar row missing a field
duplicate	0	bar ids an earlier row already carries
bars, summed	2	RECONCILED: the parts sum to the total 2
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 14. Rank input binding: STALE_INCONSISTENT. Dependency order is authored. Implementation, acceptance evidence, authority and resource readiness are unverified. No action is recommended or dispatched here.

Plan SHA-256 b1ed2073c39c7d688fcc4054efb321816035301552e4e789c1d315572e115aa7. Target rows 2; role rows 14. Existing risks and release worklog retain their own scope; no node completion is inferred.

Runtime observation coverage: 0 of 14 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.

DB0 — Unified refutation in-run
Prerequisites: None declared; this does not establish execution eligibility.
Contract: nxr_assess
Acceptance: Every numeric claim graded vs its source and cross-validated before publish -- LANDED gated
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.
DB1 — Sovereign reader
Prerequisites: None declared; this does not establish execution eligibility.
Contract: qwx_extract
Acceptance: On-protocol SQuAD 510, Hotpot 505 -- LANDED
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.
DB2 — Lexical retrieval
Prerequisites: None declared; this does not establish execution eligibility.
Contract: idf_q10
Acceptance: BM25 baseline, integer-deterministic -- LANDED
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.
DR1 — Served research API
Prerequisites: DB0 (acceptance unverified)
Contract: rapi_serve
Acceptance: An authenticated HTTPS door that runs the unified pipeline for a question and returns the graded report with per-claim verdicts; gate proves a request over the edge returns a report whose claims carry their sources and a fabricated-number fixture is refused by the same path
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.
DR2 — Gated synthesis
Prerequisites: DB0 (acceptance unverified), DB1 (acceptance unverified)
Contract: ls_synthesize
Acceptance: The sovereign model composes prose from the extractive findings; every sentence with a number or a citation is graded by nxr_assess and a discredited claim forces rewrite or STOP-INCOMPLETE, never ships; gate proves a planted false number in the synthesis is caught and the report abstains
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.
DR3 — Trained cross-encoder reranker
Prerequisites: DB2 (acceptance unverified)
Contract: nr_cross_score
Acceptance: A no-float cross-encoder scoring (query, passage) pairs, trained on the ledger's verified pairs; ACCEPT RULE declared before the run: must EXCEED the BM25 arm (321 permil on the BEIR ruler) on the same queries or it stays UNWIRED, as the semppmi and RRF arms did
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.
DR4 — Charts in the report
Prerequisites: DR2 (acceptance unverified)
Contract: mr_chart_emit
Acceptance: Numeric findings rendered as sovereign SVG charts inside the report, each chart citing the claims it plots; gate proves a chart's plotted values equal the graded claims byte-for-byte and a chart of an ungraded number is refused
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.
DR5 — Learned agent policy
Prerequisites: DB0 (acceptance unverified), DR3 (acceptance unverified)
Contract: ra_policy_step
Acceptance: A policy that chooses the next action (fetch, read, refute, stop) trained on the verified-run ledger through the no-float autograd; accept rule: higher on-protocol F1 per fetch than the hand-written loop on a held-out question set, or UNWIRED
Authored effort: 3.5. 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.
DN0 — The noticer itself
Prerequisites: None declared; this does not establish execution eligibility.
Contract: fs_deficit
Acceptance: LANDED 2026-08-20. Seeds derived from the domain's own board, outside pulled through the sovereign fetcher, findings diffed against our rows and refs, GAP PROPOSALS filed to a worklist and the frontierprop- plane. Done-rule: gate GREEN with a negative control proving a no-gap board issues ZERO fetches and ZERO proposals, and a tooth proving a run never writes the board it reads
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.
DN1 — Fleet sweep on a beat
Prerequisites: DN0 (acceptance unverified)
Contract: fs_sweep_all
Acceptance: Every /compare domain scanned on a cadence instead of one. Done-rule: a per-source pace budget that the run PRINTS and obeys, a second key-free index admitted as a source row, and a full-fleet run whose outbound request count matches rows times sources times domains exactly -- a sweep that cannot state its own request count is a load incident waiting to happen
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.
DN2 — Proposal adjudication in the hive
Prerequisites: DN0 (acceptance unverified)
Contract: fs_adjudicate
Acceptance: A seat accepts or rejects a proposal by ONE call that closes the plane row and, on accept, emits the ready-to-paste ref row with the mirror already fetched and pinned. Done-rule: an accepted proposal produces a refs row that nx_compare_refs_gate passes on first run, and a rejected one leaves the board byte-identical
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.
DN3 — Mirror bound to its own URL
Prerequisites: None declared; this does not establish execution eligibility.
Contract: crg_mirror_url_bound
Acceptance: The referee proves a mirror is a capture of the url in ITS OWN ROW, not merely that some file exists and its pin matches. Done-rule: the four 2026-08-18 charsim rows that pass today must FAIL this tooth, and every honestly-fetched row must still pass -- the bite is already sitting in the corpus
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.
DN4 — Drift checked across every register on a beat, with a verdict something can read
Prerequisites: DN3 (acceptance unverified)
Contract: rd_beat
Acceptance: The four-state organ becomes the one that RUNS, and its answer survives the run. Done-rule, all five and each independently checkable: the PROMOTED artifact prints a dynamic field in its envelope, so the state that exists in source is the state being served and the source is no longer ahead of the binary; the run truncate-writes a verdict log with a canonical last line, because a census whose verdict is never persisted has no artifact to disagree with and nothing on a beat or in a deploy-ready row can consume it; a rostered nx_refdrift_gate exists and is BITE-PROVEN in both directions, since a comparator that answered same for everything would report a permanently clean fleet and the self-verb selftest has never been mutated; the sweep covers every refs register across the fleet under a per-source pace budget the run PRINTS and OBEYS, because outbound cost is rows times registers and a sweep that cannot state its own request count is a load incident waiting to happen; and DRIFTED is RATCHETED and self-baselined on first sight rather than armed, because arming it fleet-wide would turn every domain holding an aged mirror RED in one step and teach everyone to ignore it, while DYNAMIC is NEVER counted as a defect and every DYNAMIC url is NAMED, since a count without a worklist is not actionable
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.
DR6 — Research output preserves qualitative findings and reproducible quantitative claims
Prerequisites: DR4 (acceptance unverified)
Contract: _ABSENT_:research_output_evidence_accept
Acceptance: SOURCE-BOUND PLANNING CONTRACT 2026-09-24. Compose the existing graded-claim report/chart and paper-evidence owners. A report binds question, method, input versions, qualitative observations, codebook version, numeric derivations, uncertainties, failures and reviewer decisions; no number replaces its original narrative. Demonstrate a joint display where each plotted value resolves to the source observation and derivation, and an invented value, mismatched codebook or omitted failed run is rejected or shown unresolved. Reproduce the declared result from retained inputs and distinguish independent reviewer agreement from truth or delivery acceptance. No universal photoreal percentage, invented threshold, model training or automatic paper publication. Existing documentation/Compare surfaces carry the dated result and unfinished work; effort unestimated.
Authored effort: unestimated. Executor kind: Workflow. 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 deepresearch 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.

DB0 Unified refutation in-run · cell PRESENT · gate NO-GATE-DECLARED

Contract	`nxr_assess` · executor Organ · 0 u · stands on nothing
Definition of done	Every numeric claim graded vs its source and cross-validated before publish -- LANDED gated
Matrix cell	Adversarial refutation · `runtime/nx_research_unified.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_research_unified`; declare `nx_research_unified TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

DB1 Sovereign reader · cell PRESENT · gate NO-GATE-DECLARED

Contract	`qwx_extract` · executor Organ · 0 u · stands on nothing
Definition of done	On-protocol SQuAD 510, Hotpot 505 -- LANDED
Matrix cell	Multi-hop QA reader · `runtime/nx_qwen_extract.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_qwen_extract`; declare `nx_qwen_extract TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

DB2 Lexical retrieval · cell PRESENT · gate NO-GATE-DECLARED

Contract	`idf_q10` · executor Organ · 0 u · stands on nothing
Definition of done	BM25 baseline, integer-deterministic -- LANDED
Matrix cell	Lexical retrieval (BM25) · `runtime/nx_intlog.nx` · PRESENT decl Integer-deterministic, crash-resumable, no link-rot · `runtime/nx_intlog.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_intlog`; declare `nx_intlog TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

DR1 Served research API · cell WATCHING · gate NO-GATE-DECLARED

Contract	`rapi_serve` · executor Organ · 1.5 u · stands on DB0
Definition of done	An authenticated HTTPS door that runs the unified pipeline for a question and returns the graded report with per-claim verdicts; gate proves a request over the edge returns a report whose claims carry their sources and a fabricated-number fixture is refused by the same path
Matrix cell	Served research API / product · `runtime/nx_research_api.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_research_api`; declare `nx_research_api TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

DR2 Gated synthesis · cell WATCHING · gate NO-GATE-DECLARED

Contract	`ls_synthesize` · executor Organ · 2 u · stands on DB0, DB1
Definition of done	The sovereign model composes prose from the extractive findings; every sentence with a number or a citation is graded by nxr_assess and a discredited claim forces rewrite or STOP-INCOMPLETE, never ships; gate proves a planted false number in the synthesis is caught and the report abstains
Matrix cell	LLM synthesis / reasoning · `runtime/nx_llm_synth.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_llm_synth`; declare `nx_llm_synth TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

DR3 Trained cross-encoder reranker · cell WATCHING · gate NO-GATE-DECLARED

Contract	`nr_cross_score` · executor Organ · 3 u · stands on DB2
Definition of done	A no-float cross-encoder scoring (query, passage) pairs, trained on the ledger's verified pairs; ACCEPT RULE declared before the run: must EXCEED the BM25 arm (321 permil on the BEIR ruler) on the same queries or it stays UNWIRED, as the semppmi and RRF arms did
Matrix cell	Neural reranker (cross-encoder) · `runtime/nx_neural_reranker.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_neural_reranker`; declare `nx_neural_reranker TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

DR4 Charts in the report · cell WATCHING · gate NO-GATE-DECLARED

Contract	`mr_chart_emit` · executor Organ · 1.5 u · stands on DR2
Definition of done	Numeric findings rendered as sovereign SVG charts inside the report, each chart citing the claims it plots; gate proves a chart's plotted values equal the graded claims byte-for-byte and a chart of an ungraded number is refused
Matrix cell	Multimodal report output · `runtime/nx_multimodal_report.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_multimodal_report`; declare `nx_multimodal_report TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

DR5 Learned agent policy · cell WATCHING · gate NO-GATE-DECLARED

Contract	`ra_policy_step` · executor Organ · 3.5 u · stands on DB0, DR3
Definition of done	A policy that chooses the next action (fetch, read, refute, stop) trained on the verified-run ledger through the no-float autograd; accept rule: higher on-protocol F1 per fetch than the hand-written loop on a held-out question set, or UNWIRED
Matrix cell	RL-trained research agent · `runtime/nx_research_agent.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_research_agent`; declare `nx_research_agent TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

DN0 The noticer itself · cell PRESENT · gate NO-GATE-DECLARED

Contract	`fs_deficit` · executor Organ · 0 u · stands on nothing
Definition of done	LANDED 2026-08-20. Seeds derived from the domain's own board, outside pulled through the sovereign fetcher, findings diffed against our rows and refs, GAP PROPOSALS filed to a worklist and the frontierprop- plane. Done-rule: gate GREEN with a negative control proving a no-gap board issues ZERO fetches and ZERO proposals, and a tooth proving a run never writes the board it reads
Matrix cell	Capability-gap proposals diffed against our own board · `runtime/nx_frontier_scan.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_frontier_scan`; declare `nx_frontier_scan TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

DN1 Fleet sweep on a beat · cell WATCHING · gate NO-GATE-DECLARED

Contract	`fs_sweep_all` · executor Organ · 1.5 u · stands on DN0
Definition of done	Every /compare domain scanned on a cadence instead of one. Done-rule: a per-source pace budget that the run PRINTS and obeys, a second key-free index admitted as a source row, and a full-fleet run whose outbound request count matches rows times sources times domains exactly -- a sweep that cannot state its own request count is a load incident waiting to happen
Matrix cell	Frontier noticing across every /compare domain on a beat · `runtime/nx_frontier_scan.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_frontier_scan`; declare `nx_frontier_scan TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

DN2 Proposal adjudication in the hive · cell NO-MATRIX-ROW · gate NO-MATRIX-ROW

Contract	`fs_adjudicate` · executor Organ · 1 u · stands on DN0
Definition of done	A seat accepts or rejects a proposal by ONE call that closes the plane row and, on accept, emits the ready-to-paste ref row with the mirror already fetched and pinned. Done-rule: an accepted proposal produces a refs row that nx_compare_refs_gate passes on first run, and a rejected one leaves the board byte-identical
Matrix cell	NO-MATRIX-ROW — no matrix row carries `fs_adjudicate`; the ranker reads this rung as unmapped and no cell can flip for it. Add the row and the join completes on the next publish.
Gate	the gate join needs a matrix row: no organ is named for this rung
Work log	no log row names this rung yet

DN3 Mirror bound to its own URL · cell NO-MATRIX-ROW · gate NO-MATRIX-ROW

Contract	`crg_mirror_url_bound` · executor Organ · 1 u · stands on nothing
Definition of done	The referee proves a mirror is a capture of the url in ITS OWN ROW, not merely that some file exists and its pin matches. Done-rule: the four 2026-08-18 charsim rows that pass today must FAIL this tooth, and every honestly-fetched row must still pass -- the bite is already sitting in the corpus
Matrix cell	NO-MATRIX-ROW — no matrix row carries `crg_mirror_url_bound`; the ranker reads this rung as unmapped and no cell can flip for it. Add the row and the join completes on the next publish.
Gate	the gate join needs a matrix row: no organ is named for this rung
Work log	no log row names this rung yet

DN4 Drift checked across every register on a beat, with a verdict something can read · cell WATCHING · gate NO-GATE-DECLARED

Contract	`rd_beat` · executor Organ · 1.5 u · stands on DN3
Definition of done	The four-state organ becomes the one that RUNS, and its answer survives the run. Done-rule, all five and each independently checkable: the PROMOTED artifact prints a dynamic field in its envelope, so the state that exists in source is the state being served and the source is no longer ahead of the binary; the run truncate-writes a verdict log with a canonical last line, because a census whose verdict is never persisted has no artifact to disagree with and nothing on a beat or in a deploy-ready row can consume it; a rostered nx_refdrift_gate exists and is BITE-PROVEN in both directions, since a comparator that answered same for everything would report a permanently clean fleet and the self-verb selftest has never been mutated; the sweep covers every refs register across the fleet under a per-source pace budget the run PRINTS and OBEYS, because outbound cost is rows times registers and a sweep that cannot state its own request count is a load incident waiting to happen; and DRIFTED is RATCHETED and self-baselined on first sight rather than armed, because arming it fleet-wide would turn every domain holding an aged mirror RED in one step and teach everyone to ignore it, while DYNAMIC is NEVER counted as a defect and every DYNAMIC url is NAMED, since a count without a worklist is not actionable
Matrix cell	Drift is checked on a beat across every register, with a verdict a reader can consume · `runtime/nx_refdrift.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_refdrift`; declare `nx_refdrift TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	2026-09-04 measure THE NEW ORGAN ADJUDICATED ANOTHER DETECTOR ON ITS FIRST OUTING AND THE OTHER DETECTOR WAS WRONG. The refs referee reported MIRROR-INCOMPLETE for procgen key fastnoise2 with reason json-unterminated at 6339 bytes, which would have sent a lane to re-fetch a mirror that is already whole. nx_refdrift fetched that url twice and both fetches returned EXACTLY the stored 6339 bytes, so the mirror is complete and the finding is a FALSE POSITIVE. The mechanism is a format inferred from the wrong signal: the body is a raw GitHub markdown README stored under a .raw name, and the classifier read it as JSON and then complained the JSON never closed. This is a SECOND false-positive class in that detector, distinct from the HTML5 optional-end-tag class measured at 941 permil earlier the same week. A re-fetch returning the identical byte count is the control that separates a truncation from a complete body, and it is the same control that settled the earlier class. 2026-09-04 measure THE FOUR-STATE VERDICT EARNED ITSELF IMMEDIATELY. Live over procgen.refs: refs_total 22 checked 13 match 10 drifted 2 unreachable 0 dynamic 1 sum 13, with PARTIAL declared because the run is a prefix of the file. The three-state version shipped one hour earlier would have reported that dynamic row as DRIFTED and sent someone to re-pin a page that simply varies per request. A detector with false positives is worse than none because it teaches the fleet to ignore it, and the second fetch that separates the two costs one request. 2026-09-04 queue THE UNCHECKED REMAINDER IS DECLARED, NOT HIDDEN. 13 of 22 rows were checked with max set to 40, so the stop was not the bound: the other 9 rows did not yield a readable mirror and were skipped without being counted as clean. That skip is currently silent in the per-row output and it should be its own named state beside MATCH, DRIFTED, UNREACHABLE and DYNAMIC, because a row that was never examined must never read as one that passed. That is the next correction to this organ and it is the same law it already applies to its PARTIAL line.

DR6 Research output preserves qualitative findings and reproducible quantitative claims · cell NO-MATRIX-ROW · gate NO-MATRIX-ROW

Contract	`_ABSENT_:research_output_evidence_accept` · executor Workflow · unestimated u · stands on DR4, researchreach:RR15 (UNRESOLVED: no such rung on this plan)
Definition of done	SOURCE-BOUND PLANNING CONTRACT 2026-09-24. Compose the existing graded-claim report/chart and paper-evidence owners. A report binds question, method, input versions, qualitative observations, codebook version, numeric derivations, uncertainties, failures and reviewer decisions; no number replaces its original narrative. Demonstrate a joint display where each plotted value resolves to the source observation and derivation, and an invented value, mismatched codebook or omitted failed run is rejected or shown unresolved. Reproduce the declared result from retained inputs and distinguish independent reviewer agreement from truth or delivery acceptance. No universal photoreal percentage, invented threshold, model training or automatic paper publication. Existing documentation/Compare surfaces carry the dated result and unfinished work; effort unestimated.
Matrix cell	NO-MATRIX-ROW — no matrix row carries `research_output_evidence_accept`; the ranker reads this rung as unmapped and no cell can flip for it. Add the row and the join completes on the next publish.
Gate	the gate join needs a matrix row: no organ is named for this rung
Work log	no log row names this rung yet

dossiers=14 cells_joined=11 no_matrix_row=3 (sum 14) gates_recorded=0 gates_unrecorded=0 no_gate_declared=11 (sum 11 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. 19 feature rows mapped of 25 matrix rows; 6 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/deepresearch.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 700–1000, distinctiveness 50–350 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 0 (a fact of the data: two players that close are that close) · data spread 642 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 700–1000 · distinctiveness 50–350 · breadth 300–550 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 0 · unresolved labels 0 · mark overlaps 1 (a fact of the data: two players that close are that close) · data spread 363 permil of the plot

Nishi Perplexity DeepResearcher Tongyi deep-searcher
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 4 (a fact of the data: two players that close are that close) · data spread 518 permil of the plot

Player	Quadrant	centrality	distinctiveness	breadth	depth	momentum	Rows held	Days on spine	First seen
Nishi	Unconventional	733	272	421	727	0 since 2026-09-23	11	10	2026-09-15
Perplexity	Aspirational	933	224	508	743	0 since 2026-09-23	13	10	2026-09-15
DeepResearcher	Mainstream	916	104	421	666	0 since 2026-09-23	12	10	2026-09-15
Tongyi	Mainstream	950	115	438	641	0 since 2026-09-23	13	10	2026-09-15
deep-searcher	Peripheral	866	121	385	666	0 since 2026-09-23	11	10	2026-09-15

Day	Matrix rows reviewed	Players recorded
2026-09-15	25	5
2026-09-16	25	5
2026-09-17	25	5
2026-09-18	19	5
2026-09-19	19	5
2026-09-20	19	5
2026-09-21	19	5
2026-09-22	19	5
2026-09-23	19	5
2026-10-08	19	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 100, unassessed 0. Authored assessments are not verified outcomes.

Capability	Nishi	Perplexity	DeepResearcher	Tongyi	deep-searcher
Web fetch and bank pipeline Measured: `rf_fetch_bank` exists in `runtime/nx_research_engine.nx`, verified at emit. RAG fetch loop; all fetch the live web rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them Adoption: `BUILT-UNPROMOTED` — PARTIAL: compiled, never promoted to the serving root: /api/promote it. not adopted: compiled, never promoted to the serving root: /api/promote it Pros Implementing source symbol found; runtime acceptance separateCons not adopted yet: compiled, never promoted to the serving root: /api/promote it	●	●	●	●	●
Lexical retrieval (BM25) Measured: `idf_q10` exists in `runtime/nx_intlog.nx`, verified at emit. Sparse retrieval baseline 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=9 nonval=7` — 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)	●	●	●	●	●
Dense vector retrieval Measured: `nx_vec_index` exists in `runtime/nx_vec_index.nx`, verified at emit. deep-searcher leads (Milvus / Zilliz vector DB) [deepsearcher] Adoption: `SOURCE-ONLY` — PARTIAL: source exists, never compiled: /api/build it. not adopted: source exists, never compiled: /api/build it Pros Implementing source symbol found; runtime acceptance separateCons behind deep-searcher (leads); not adopted yet: source exists, never compiled: /api/build it	●	●	●	●	◉
Distributional embeddings Measured: `nx_distrib_embed` exists in `runtime/nx_distrib_embed.nx`, verified at emit. Co-occurrence semantic vectors rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them Adoption: `SOURCE-ONLY` — PARTIAL: source exists, never compiled: /api/build it. not adopted: source exists, never compiled: /api/build it Pros Implementing source symbol found; runtime acceptance separateCons not adopted yet: source exists, never compiled: /api/build it	●	●	●	●	●
PPMI semantic matrix Measured: `nx_semppmi_build` exists in `runtime/_hdl_build/nx_semppmi_build.nx`, verified at emit. Ours explicit sparse; theirs neural rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them 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 Perplexity, DeepResearcher, Tongyi, deep-searcherCons 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)	●	◐	◐	◐	◐
Trained embeddings Measured: `nx_embed_train` exists in `runtime/_hdl_build/nx_embed_train.nx`, verified at emit. Ours PPMI-factorization toy; theirs neural leads rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them Adoption: `REGISTERED-UNAUTHORISED` — PARTIAL: registered, no cap ever minted; no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). not adopted: registered, no cap ever minted; no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked) Pros Implementing source symbol found; runtime acceptance separateCons behind Perplexity (leads), DeepResearcher (leads), Tongyi (leads), deep-searcher (leads); not adopted yet: registered, no cap ever minted; no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked)	●	◉	◉	◉	◉
Multi-hop QA reader Measured: `qwx_extract` exists in `runtime/nx_qwen_extract.nx`, verified at emit. Pretrained sovereign Qwen 1-shot + s0-window opt-in: on-protocol SQuAD 510 and Hotpot 505 (above mech-oracle 368) [yang2018hotpot] vs mech 250; MuSiQue 122 ties the composition wall [trivedi2022musique] (NAS-measured, multi-alias). A 2-shot surface-form variant won +104 on the single-gold standalone bench but did NOT transfer to the multi-alias on-protocol metric (SQuAD 427 vs 510) and was reverted (confirmed by nx_reader_ruler_adversary_gate 9/9). Evidence-grounded: SQuAD human F1 86.8 per arxiv:1606.05250 [rajpurkar2016]; published SQuAD v1.1 SOTA = F1 932 (BERT, arxiv:1810.04805) [devlin2018] VERIFIED via nx_research_verify -- our reader 510 = 55 percent of SOTA, which still leads Adoption: `LIB-WIRED importers=2 nonval=1` — fully adopted (top of its ladder). dossier Pros Implementing source symbol found; runtime acceptance separate; fully adopted on the estate ladderCons behind Perplexity (leads), DeepResearcher (leads), Tongyi (leads)	●	◉	◉	◉	●
QA F1 / EM scoring Measured: `qs_f1` exists in `runtime/nx_qa_score_lib.nx`, verified at emit. SQuAD-style measured evaluation 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 none on the measured axes (rival marks are documented presence, not depth)	●	●	●	●	●
Findings ledger (compounding memory) Measured: `rl_upsert` exists in `runtime/nx_research_ledger.nx`, verified at emit. Persistent provenance ledger rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them Adoption: `BUILT-UNPROMOTED` — PARTIAL: compiled, never promoted to the serving root: /api/promote it. not adopted: compiled, never promoted to the serving root: /api/promote it Pros Implementing source symbol found; runtime acceptance separate; ahead of deep-searcherCons not adopted yet: compiled, never promoted to the serving root: /api/promote it	●	●	●	●	◐
Corpus semantic retrieval Measured: `nx_lib_semantic` exists in `runtime/nx_lib_semantic.nx`, verified at emit. deep-searcher leads rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them Adoption: `SOURCE-ONLY` — PARTIAL: source exists, never compiled: /api/build it. not adopted: source exists, never compiled: /api/build it Pros Implementing source symbol found; runtime acceptance separateCons behind deep-searcher (leads); not adopted yet: source exists, never compiled: /api/build it	●	●	●	●	◉
RL-trained research agent Open — watching `runtime/nx_research_agent.nx` : `ra_policy_step`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. DeepResearcher / Tongyi GRPO-trained lead; Nishi none [shao2024grpo] dossierwatching ra_policy_step Pros none measured yetCons behind Perplexity, DeepResearcher (leads), Tongyi (leads); open contract; expected source declaration not established	○	●	◉	◉	○
LLM synthesis / reasoning Open — watching `runtime/nx_llm_synth.nx` : `ls_synthesize`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. Perplexity leads (polished synthesis); Nishi extractive only rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them dossierwatching ls_synthesize Pros none measured yetCons behind Perplexity (leads), DeepResearcher, Tongyi, deep-searcher; open contract; expected source declaration not established	○	◉	●	●	●
Adversarial refutation Measured: `nxr_assess` exists in `runtime/nx_research_unified.nx`, verified at emit. DeepResearcher cross-validates across sources (leads). Nishi now has nx_claim_verify (single) + nx_research_verify (batch STAGE) -- autonomous: fetch a source over sovereign TLS then confirm/discredit/flag-unverifiable a claim vs primary evidence, never fabricating (LIVE-PROVEN: claim 3/3, batch 4/4, and grounded the SQuAD SOTA 932 above) + metric-ruler adversary (nx_reader_ruler_adversary_gate 9/9). Single-source numeric-grounding (catches fabrication/alteration) vs multi-source cross-validation. NOW AUTO-INVOKED in-run (2026-07-14): nx_research_unified (nxr_assess) grades every numeric claim vs its cited source and GATES PUBLISH -- a discredited/fabricated number forces CONTINUE or STOP-INCOMPLETE, never ships; gate-proven nx_research_verify_synth_gate 4/4 + nx_research_unified_gate 5/5 + nx_research_refute_gate 7/7. So PRESENT (auto-invoked, gated). AND multi-source cross-validation is NOW ALSO auto-invoked (2026-07-14b): cvx_answer_verify (nx_research_crossval) inside nxr_assess corroborates each claim across ALL sources with the independence discipline -- >=2 distinct origins required, same-origin echo never counts, conflict surfaced not averaged; crossval gate 10/10 incl a LIVE 3-host proof (GPT-3 175B CORROBORATED across arxiv.org + api.datacite.org while a rate-limited semanticscholar stub was absorbed as unverifiable, not faked). DeepResearcher's agentic multi-source loop still leads on breadth, so NOT exceed [deepresearcher2025] Adoption: `LIB-GATE-ONLY importers=3` — PARTIAL: imported only by validation organs (gates, tests, benches): wire it into a shipping program. dossiernot adopted: imported only by validation organs (gates, tests, benches): wire it into a shipping program Pros Implementing source symbol found; runtime acceptance separate; ahead of deep-searcherCons behind DeepResearcher (leads); not adopted yet: imported only by validation organs (gates, tests, benches): wire it into a shipping program	●	●	◉	●	○
Neural reranker (cross-encoder) Open — watching `runtime/nx_neural_reranker.nx` : `nr_cross_score`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. Trained rerank; Nishi lexical + sparse only rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them dossierwatching nr_cross_score Pros none measured yetCons behind Perplexity, DeepResearcher, Tongyi, deep-searcher; open contract; expected source declaration not established	○	●	●	●	◐
Multimodal report output Open — watching `runtime/nx_multimodal_report.nx` : `mr_chart_emit`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. Perplexity leads (charts / images); Nishi text only rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them dossierwatching mr_chart_emit Pros none measured yetCons behind Perplexity (leads); open contract; expected source declaration not established	○	◉	○	○	○
Served research API / product Open — watching `runtime/nx_research_api.nx` : `rapi_serve`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. Perplexity leads (product API); Nishi organ-only [perplexity-api] [scoutforge-methodology-20260919] [scoutforge-remnus-20260919] dossierwatching rapi_serve Pros none measured yetCons behind Perplexity (leads), DeepResearcher, Tongyi, deep-searcher; open contract; expected source declaration not established	○	◉	◐	◐	●
Sovereign -- own substrate, TLS, index; zero OpenAI / cloud dep; 0 dollar per query Source present; authored stronger claim: `ss_fold_cp` in `runtime/nx_seg_store.nx`, verified at emit. All use hosted LLMs; Tongyi open-weights partial [tongyi2025]; Nishi fully sovereign Adoption: `LIB-WIRED importers=498 nonval=372` — fully adopted (top of its ladder). Pros Source symbol found; stronger capability claim is authored; ahead of Perplexity, DeepResearcher, Tongyi, deep-searcher; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	◐	○
Integer-deterministic, crash-resumable, no link-rot Source present; authored stronger claim: `idf_q10` in `runtime/nx_intlog.nx`, verified at emit. Reproducible; peers are non-deterministic LLM pipelines Adoption: `LIB-WIRED importers=9 nonval=7` — fully adopted (top of its ladder). dossier Pros Source symbol found; stronger capability claim is authored; ahead of Perplexity, DeepResearcher, Tongyi, deep-searcher; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
Capability-gap proposals diffed against our own board Source present; authored stronger claim: `fs_deficit` in `runtime/nx_frontier_scan.nx`, verified at emit. LANDED 2026-08-20, nx_frontier_scan_gate GREEN 14 of 14 with two teeth BITE-PROVEN. The organ derives its query seeds from a /compare domain's OWN data (the .axes frontier-keyword field, the .matrix row labels), pulls the outside through the sovereign fetcher over sources that are themselves data [arxiv-api], and emits GAP PROPOSALS naming the row each candidate challenges plus its fetched mirror. THE FOUR RIVAL COLUMNS ARE CODED 0 AS A CATEGORY STATEMENT AND NOT AS A FEATURE AUDIT: all four are answer engines that respond to a question a human asked, none maintains a capability board, so there is no comparable surface to measure them on and none is fabricated here. It emits proposals and NEVER admits one -- the measured law is that no margin threshold makes auto-declaration safe, and a completion signal that keys on a name rewards writing the name. Adoption: `RUN-BY:clock` — fully adopted (top of its ladder). dossier Pros Source symbol found; stronger capability claim is authored; ahead of Perplexity, DeepResearcher, Tongyi, deep-searcher; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
Frontier noticing across every /compare domain on a beat DARK — `runtime/nx_frontier_scan.nx` EXISTS but does not declare `fs_sweep_all`: 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. OPEN. One domain is on a daily beat today (cron.reg, 07:35) at 12 outbound requests per run, because outbound cost is rows times sources. A fleet sweep over 58 domains at the same settings is roughly 700 requests a day, which is a rate-limit and host-load decision before it is a scheduling one, so the rung must admit a second key-free scholarly index [openalex-api] and a per-source pace budget before it can run wide. Rival columns unmeasured for the category reason given on the row above. [arxiv_dr_202609] dossierwatching fs_sweep_all Pros none measured yetCons open contract; expected source declaration not established	○	○	○	○	○
Citation mirror provably bound to the URL it claims Source present; authored stronger claim: `crg_prov_state` in `runtime/nx_compare_refs_gate.nx`, verified at emit. LANDED 2026-08-20, nx_compare_refs_gate GREEN 18 of 18. THE CONTRACT WAS DECLARED AS crg_mirror_url_bound AND THE SHIPPED SYMBOL IS crg_prov_state -- the rename is recorded here rather than left to read as a moved goalpost: the contract named a property, the function names the three-state answer. The binding is made at FETCH time, the only moment both facts are in hand -- nx_research_fetch appends a row carrying epoch, requested-url, mirror, sha256, bytes and redirect-hops to knowledge/status/fetch_provenance.jrnl, and this gate joins every declared mirror against it. THREE STATES because two would lie: PROVEN, MISMATCH (this mirror under a DIFFERENT url), and UNPROVEN (no journal row at all, which is every mirror captured before the journal existed and is NOT a defect). Live 2026-08-20: jrnl_rows 29, proven 4, mismatch 0, unproven 617, declared_mirrors 621, and the partition SUMS. MISMATCH is ARMED at zero; UNPROVEN is RATCHETED and self-baselined at 617, because arming it fleet-wide would have turned 68 domains RED in one step and taught everyone to ignore the gate. Every unproven mirror is NAMED in knowledge/status/refs_provenance_unproven.txt, because a count without a worklist is not actionable. Bite-proven in BOTH directions on fixtures assembled at runtime, with a tooth asserting the fixtures actually parsed before any verdict is read off them. RESIDUAL, STATED IN THE GATE VERDICT ITSELF AND CARRIED AS THE ROW BELOW: provenance proves the mirror is OF the url, never that it SUPPORTS the claim. THE HOLE IT CLOSED, kept here so the fix keeps its reason: it is the exact hole the 2026-08-18 metahuman episode fell through. Four references were appended whose URLs were never opened; three carry mirror and pin both absent, and the fourth points at a page fetched three days earlier from a DIFFERENT url -- and the referee passed the file 9 of 9, because it proves the mirror EXISTS and the pin EQUALS the filehash of that mirror, never that the mirror is a capture of the row's own url. A PIN PROVES THE BYTES DID NOT CHANGE, NEVER THAT THEY ARE THE RIGHT DOCUMENT. The same blindness has a second cause now closed: until 2026-08-20 the TLS-1.2 fetch leg truncated bodies near 256 KiB while printing a clean SAVED, so a partial page could be pinned as whole. Rival columns unmeasured for the category reason given two rows above. Adoption: `GATE:LIVE trial=-` — fully adopted (top of its ladder). Pros Source symbol found; stronger capability claim is authored; ahead of Perplexity, DeepResearcher, Tongyi, deep-searcher; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
Citation mirror provably SUPPORTS the claim it backs DARK — `runtime/nx_compare_refs_gate.nx` EXISTS but does not declare `crg_claim_supported`: 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. OPEN, and it is the half that fetch-time provenance CANNOT close -- named as its own contract rather than left implied, because the row above could easily be read as having ended the class. Provenance proves the mirror is a capture of the url in its row. It says NOTHING about whether that document contains the claim the row cites it for, and on 2026-08-20 that second question had to be answered BY HAND over four mirrors and 22 claims: PRESENT 6, PARTIAL 2, ABSENT 13. Every one of the 13 sat behind a real file with a valid pin, and after this rung ships they would still sit behind a real file with a valid pin AND a correct provenance row. The done-rule is a checker that takes a row grounds field and its mirror and returns SUPPORTED, UNSUPPORTED or UNDECIDABLE with the witnessing byte offsets, ratcheted the same way and armed on nothing until it has a calibrated false-positive rate on the existing corpus. Rival columns unmeasured for the category reason given three rows above. watching crg_claim_supported Pros none measured yetCons open contract; expected source declaration not established	○	○	○	○	○
Frontier sources that are not scholarly indexes -- vendor release notes and conference programmes DARK — `runtime/nx_frontier_scan.nx` EXISTS but does not declare `fs_src_page`: 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. OPEN, and the reason it is open is a COVERAGE BOUNDARY, not a scoring failure. THE LOOP CANNOT SEE THESE, IT DID NOT MERELY MISS THEM. The two wired sources are arXiv and HN Algolia. Measured against the four references a human added to charsim on 2026-08-18: metahuman57 is an Epic Developer Community release note and rtxhair26 is a GDC schedule session page -- NEITHER INDEX CARRIES EITHER KIND OF PAGE AT ALL; cc5-reallusion is a trade-press article, reachable through HN only on the condition that some human happened to submit it, which is not a capability of the source; sig26-avatars is a SIGGRAPH changelog index page, itself unreachable, though the papers it lists are on arXiv under different citations. SO THE HONEST DENOMINATOR IS ZERO: of the four cited urls, ZERO were in-principle reachable from the sources wired, and the replays 0-of-4 therefore means the loop missed 0 of the 0 it could see. THE MECHANISM ITSELF WORKS WHERE ITS SOURCES CAN SEE -- the 2026-08-20 v2 replay surfaced Show HN SwitchLight, a browser relighting product, against the SKIN full-PBR-and-normal-map-lighting row, unaided and on-topic. PRE-DECLARED ACCEPT RULE, WRITTEN BEFORE THE SOURCE EXISTS SO IT CANNOT BE RATIONALIZED AFTERWARDS: wiring a vendor and release-note source kind is accepted only if replaying the 2026-08-18 PRE-STATE board surfaces the MetaHuman-class surface-layer gap unaided, as at least one proposal whose url is a vendor release note challenging a SKIN, HAIR, MESH or ANIM row, and only if that proposal mirror carries a row in knowledge/status/fetch_provenance.jrnl binding it to that url. THE SECOND CONDITION IS NOT DECORATION: the same audit proved Epic SILENTLY REPOINTED the metahuman-5-7 slug to the 5.8 notes, so a vendor page can change identity underneath a citation, and a vendor source without fetch-time provenance would rebuild the exact hole this lane just spent a day retracting. Rival columns unmeasured for the category reason given four rows above. watching fs_src_page Pros none measured yetCons open contract; expected source declaration not established	○	○	○	○	○
Citation still served: does the URL still serve the bytes we pinned Measured: `rd_same` exists in `runtime/nx_refdrift.nx`, verified at emit. GROUNDED 2026-09-06, LATER THE SAME DAY: rd_same NOW EXISTS in nx_refdrift.nx and is LIVE (promoted 4a87c99c, 65360 B). It was extracted as the FOUR-STATE CLASSIFIER -- MATCH, DRIFTED, DYNAMIC, UNREACHABLE -- and NOT as a byte comparator, because rp_same in nx_rowparse_lib already was the byte comparator and writing a second one would have been the duplicate-ruler defect. THE MISSING ENTRY POINT WAS NEVER THE MISSING CAPABILITY: what had no name was the DECISION wrapping a primitive that already existed. The extraction did not widen anything -- the caller still performs both fetches and both slurps under exactly its prior guards -- and NEUTRALITY IS PROVEN RATHER THAN ASSERTED: nx_contentdiff live-vs-candidate reports runs_scanned=260 lost_from_live=0 lost_permil=0 symmetry=ONE-WAY-GAIN verdict=GREEN, a strict superset. nx_domain_admit now reports ungrounded=0 for this domain. THE EXCEED CODE IS LEFT AT 0 DELIBERATELY AND IS THE OWNING LANE'S TO RE-ASSERT: this lane withdrew a Best it could not ground and will not restore one it has not measured. Earlier note preserved: THE SYMBOL WAS UNGROUNDED, SO THE EXCEED CLAIM WAS WITHDRAWN UNTIL THE CONTRACTED ENTRY POINT EXISTED. nx_domain_admit reports UNGROUNDED symbol rd_same not word-boundary-present in runtime/nx_refdrift.nx. A grep over buildroot/runtime with corpus_complete=1 finds func rd_same in EXACTLY ONE file and it is nx_rebuild_drain.nx, an unrelated organ sharing the rd_ prefix -- so a verification that greps the symbol NAME rather than the organ would have CONFIRMED this claim and closed the finding. The complete outline of nx_refdrift.nx reports 64 declarations, whose functions are rd_slurp, rd_put2, rd_ymd, rd_basename, rd_sha, rd_recite_row and main, with NO comparator among them: THE FOUR-STATE COMPARATOR IS REAL AND RUNS INLINE IN main, so the capability genuinely exists and its contracted entry point was simply never written. NOTHING ABOUT THE ORGAN OR ITS MEASURED FIRST RUN IS DISPUTED, and the original note is preserved verbatim below; only the symbol field and the exceed code are corrected, because a published Best on a symbol that is not in the organ it names is the darkness class this estate forbids, and it fails in the flattering direction. The extraction is carried as rung RR13 on /compare/researchreach with a neutrality proof by nx_behaveprobe. ORIGINAL NOTE, UNEDITED, FOLLOWS: LANDED 2026-09-04, and it closes the residual the refs referee states about ITSELF, quoted from that verdict: fetch-time provenance proves the mirror is OF the url, it does NOT prove the url still serves those bytes. THIS IS A THIRD AND ORTHOGONAL AXIS TO THE TWO ROWS ABOVE, not a refinement of either: bound-to-the-url is a claim about IDENTITY, supports-the-claim is a claim about CONTENT, and this is a claim about TIME. A pin proves the bytes on OUR disk did not change and nothing in the register ever asked whether the SOURCE changed, so a citation can rot in complete silence with every pin check still green -- a claim with an expiry date nobody recorded. Measured the hard way: two pinned standards had MOVED upstream since they were fetched and a seat found out only by re-fetching them for an unrelated reason. IT COMPARES BYTES AND NOT HASHES, DELIBERATELY: the referee already proves pin equals filehash of the mirror, so the mirror IS the pinned bytes, and re-fetching and byte-comparing answers the drift question exactly with no second hash implementation to drift from the first. ONE RULER. THE THIRD STATE IS THE FINDING, NOT A POLISH PASS. On the first live run one url returned THREE DIFFERENT HASHES ACROSS THREE FETCHES AT AN IDENTICAL 106703 BYTES, so a size check calls that page stable and a hash check calls it drifted and BOTH ARE WRONG -- it varies against itself per request and is UNJUDGEABLE by any byte compare. What exposed it is that one of the two rows first reported DRIFTED had been pinned FORTY MINUTES EARLIER. So the organ fetches TWICE and the second fetch is the discriminator, and it reports four named states -- MATCH, DRIFTED, UNREACHABLE, DYNAMIC -- whose counts print as a partition that SUMS, with UNREACHABLE and DYNAMIC each carrying an explicit NOT A DRIFT VERDICT marker so neither can ever read as clean. Folding DYNAMIC into DRIFTED would have shipped a false-positive generator, which this estate ranks as worse than no detector at all. BOUNDED BY CONSTRUCTION AND SAYING SO: one refs file and a maximum rows per run, both from argv, and the envelope prints refs_total and checked and PARTIAL with a named refusal to read a capped run as a clean sweep. STATE MEASURED 2026-09-04, NOT ASSERTED: nx_catalog reports SOURCE 15396 B, BUILT 33672 B, PROMOTED 33672 B, REGISTERED, AUTHORISED, INVOKED, verdict LIVE, and the source carries 7 selftest teeth counted in it including a comparator negative control where a length-only difference must read as drift and a fixture-reached-the-condition tooth. FIRST LIVE RUN over ecosystem.refs, reported by the lane that built the organ and NOT re-run here: refs_total 6, checked 6, match 4, drifted 2, unreachable 0, sum 6, PARTIAL 0. THREE RESIDUALS NAMED HERE RATHER THAN LEFT FOR THE NEXT READER, all of them the next rung and none of them a claim made by this row. One, the DYNAMIC state is SOURCE-LANDED AND NOT PROVEN SERVED: the built and promoted artifacts carry mtime 1788528974 and 1788528975 while the source reads 1788529272, so the source is 298 seconds AHEAD of the binary and that first run envelope carries no dynamic field, which means the four-state organ is the SOURCE today and the three-state organ is what runs until it is rebuilt and promoted. Two, it writes NO durable verdict log: knowledge/status holds only its magic-ratchet baseline over 5267 files with corpus_complete=1, so there is no artifact for anything to disagree with and no beat or deploy-ready row can consume its verdict. Three, its proof is a self-verb selftest and NOT a rostered gate: nx_refdrift_gate is ABSENT with no source and no binary, so nothing has bite-proven this detector in both directions. Rival columns are coded 0 as the same CATEGORY STATEMENT the four rows above use and not as a feature audit: all four are answer engines that cite live urls at answer time and none maintains a pinned mirror register, so there is no comparable surface to drift-check and nothing is fabricated here. 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 Perplexity, DeepResearcher, Tongyi, deep-searcherCons 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)	●	○	○	○	○
Drift is checked on a beat across every register, with a verdict a reader can consume DARK — `runtime/nx_refdrift.nx` EXISTS but does not declare `rd_beat`: 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. OPEN CONTRACT for rung DN4. The four-state organ is LIVE and proven on real data, separating MATCH from DRIFTED from UNREACHABLE from DYNAMIC and declaring PARTIAL when a run is a prefix, but it runs only when a seat types it. Today a pin that has moved is discovered by accident, which is how two industry standards were found stale. Done when a beat rotates every refs file, writes a truncate-written status artifact whose LAST line is the canonical verdict, and a row that was never examined is its own named state rather than reading as one that passed dossierwatching rd_beat Pros none measured yetCons 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	17	rows carrying a Yes, Best or Part in a rival column
cited	6	a reference mark resolves to a pinned mirror
uncited	11	an observation read, badged until a reference lands
rows, summed	17	RECONCILED: the parts sum to the total 17

Delivery and evidence

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

Risk register

Risk	Likelihood x impact	Mitigation
Synthesis that reads well hides a wrong number better than extractive text did	likely x high	DR2 is graded sentence by sentence by the existing referee; fluency never bypasses the gate.
A learned reranker or policy that wins on the training ruler and loses held-out is an illusion of progress	possible x high	DR3 and DR5 carry pre-declared accept rules on held-out sets; a miss stays UNWIRED with the number published.

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
2026-09-04	DN4	measure	THE NEW ORGAN ADJUDICATED ANOTHER DETECTOR ON ITS FIRST OUTING AND THE OTHER DETECTOR WAS WRONG. The refs referee reported MIRROR-INCOMPLETE for procgen key fastnoise2 with reason json-unterminated at 6339 bytes, which would have sent a lane to re-fetch a mirror that is already whole. nx_refdrift fetched that url twice and both fetches returned EXACTLY the stored 6339 bytes, so the mirror is complete and the finding is a FALSE POSITIVE. The mechanism is a format inferred from the wrong signal: the body is a raw GitHub markdown README stored under a .raw name, and the classifier read it as JSON and then complained the JSON never closed. This is a SECOND false-positive class in that detector, distinct from the HTML5 optional-end-tag class measured at 941 permil earlier the same week. A re-fetch returning the identical byte count is the control that separates a truncation from a complete body, and it is the same control that settled the earlier class.
2026-09-04	DN4	measure	THE FOUR-STATE VERDICT EARNED ITSELF IMMEDIATELY. Live over procgen.refs: refs_total 22 checked 13 match 10 drifted 2 unreachable 0 dynamic 1 sum 13, with PARTIAL declared because the run is a prefix of the file. The three-state version shipped one hour earlier would have reported that dynamic row as DRIFTED and sent someone to re-pin a page that simply varies per request. A detector with false positives is worse than none because it teaches the fleet to ignore it, and the second fetch that separates the two costs one request.
2026-09-04	DN4	queue	THE UNCHECKED REMAINDER IS DECLARED, NOT HIDDEN. 13 of 22 rows were checked with max set to 40, so the stop was not the bound: the other 9 rows did not yield a readable mirror and were skipped without being counted as clean. That skip is currently silent in the per-row output and it should be its own named state beside MATCH, DRIFTED, UNREACHABLE and DYNAMIC, because a row that was never examined must never read as one that passed. That is the next correction to this organ and it is the same law it already applies to its PARTIAL line.

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. The coverage above records source-symbol presence. Stronger comparisons are authored claims, not measured runtime superiority or accepted outcomes. A watch contract names an expected source symbol; finding it changes source discovery only. Qualification, deployment and runtime acceptance require their own evidence.

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

Every compare carries this layer. Declare knowledge/compare/deepresearch.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 deepresearch, 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 deepresearch.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	2
scanned	2
fetched	2
reused	0
failed	0
named	2
candidates	2
mentions	2
capped	0

Rank	Rival	Seeds	Mentions	First seed	Kind	Link
1	Perplexity	1	1	col0-perplexity	named	https://docs.perplexity.ai/getting-started/overview
2	deep-searcher	1	1	col3-deep	named	https://github.com/zilliztech/deep-searcher

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 24 capabilities: 18 exclusive to it and 6 shared with other boards. Shared rows on this page: 6 = home here 1 + home elsewhere 2 + ambiguous 0 + undecided 3 + refused 0 (partition 6); undecided = unmapped 0 + unproven 3 + unevidenced 0; homonyms 0; since the previous census: new 0, resolved 0. Census FRESH age_s=6165.

Capability	Organ	Boards grading it	Home	Rule	Evidence	Standing on this board
idf_q10	`nx_intlog.nx`	search+, webscraping, deepresearch*+, researchreach+	deepresearch	rung	plan	home here
nx_vec_index	`nx_vec_index.nx`	search, deepresearch	—	none	unproven	undecided
ss_fold_cp	`nx_seg_store.nx`	search, webscraping, deepresearch, automation	—	none	unproven	undecided
rf_fetch_bank	`nx_research_engine.nx`	deepresearch, instrument	—	none	unproven	undecided
crg_prov_state	`nx_compare_refs_gate.nx`	deepresearch, researchreach*	researchreach	rung	plan	home elsewhere adopt: `@xrow researchreach\|crg_prov_state\|home by rung (fleet duplicate census)`
rd_same	`nx_refdrift.nx`	deepresearch, researchreach*	researchreach	rung	plan	home elsewhere adopt: `@xrow researchreach\|rd_same\|home by rung (fleet duplicate census)`

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 1395270 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.

[perplexity-api] Perplexity. API Platform overview -- Router, Agent, Search and Embeddings APIs (real-time web-grounded answers). Vendor documentation. publisher · read in our library knowledge/fetched/cmp_deepresearch_perplexity.html · pin hae63a992df31746dde53a7ba823cfe0479d11e091814e8fcaaf8dd86c7bc0e46 · accessed 2026-08-18 · vendor-docGrounds: The Perplexity column: Best on "Served research API / product", "LLM synthesis / reasoning" and "Multimodal report output" -- the documented product API is the served-answer-engine capability those codes describe, and it is a hosted LLM service, which is the foil for the "Sovereign -- own substrate, TLS, index; zero OpenAI / cloud dep" exceed row.
[deepresearcher2025] Zheng, Y., Fu, D., Hu, X., Cai, X., Ye, L., Lu, P., Liu, P. DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environments. arXiv:2504.03160, April 2025. publisher · read in our library knowledge/fetched/cmp_deepresearch_deepresearcher2025.html · pin hfc6c302637b8af3e2c88af6cc438c81002200150b0c2641ad1fb7b62fc4f0549 · accessed 2026-08-18 · published-paperGrounds: The DeepResearcher column: Best on "RL-trained research agent" and "Adversarial refutation" -- the abstract describes end-to-end RL training of deep-research agents in real web environments and reports emergent cross-validation of information across multiple sources, which is exactly the multi-source breadth the row note concedes still leads ours.
[tongyi2025] Tongyi DeepResearch Team, Li, B., Zhang, B. et al. Tongyi DeepResearch Technical Report. arXiv:2510.24701, October 2025 (30.5 billion total parameters, 3.3 billion activated per token; model, framework and solutions open-sourced). publisher · read in our library knowledge/fetched/cmp_deepresearch_tongyi2025.html · pin h24ac13b5b8fafe85408e0b6c254151046e31913659f1c4fc1a2afa52bf3b35e8 · accessed 2026-08-18 · published-paperGrounds: The Tongyi column: Best on "RL-trained research agent" (agentic mid-training plus post-training) and Part on the "Sovereign" row ("Tongyi open-weights partial") -- the report states the model is open-sourced, which is why its column is the only rival with a non-zero code on that exceed row.
[deepsearcher] Zilliz. deep-searcher -- open source deep research alternative to reason and search on private data, written in Python (GitHub repository, zilliztech). publisher · read in our library knowledge/fetched/cmp_deepresearch_deepsearcher.html · pin hf906e0ee916636ecf7db998d4a955ff020af6a15cb02bb76220a32c1f768d46b · accessed 2026-08-18 · source-readGrounds: The deep-searcher column: Best on "Dense vector retrieval" and "Corpus semantic retrieval" (the note: Milvus / Zilliz vector DB) -- the repository is the vector-database-backed private-data research agent those codes describe.
[rajpurkar2016] Rajpurkar, P., Zhang, J., Lopyrev, K., Liang, P. SQuAD: 100,000+ Questions for Machine Comprehension of Text. EMNLP 2016; arXiv:1606.05250. Human performance F1 86.8 percent (abstract). publisher · read in our library knowledge/fetched/cmp_deepresearch_rajpurkar2016.html · pin hec5af640421c7c422bd13b49c0d2deac62a706d71047fad2304930cc756f48ce · accessed 2026-08-18 · datasetGrounds: The "Multi-hop QA reader" and "QA F1 / EM scoring" rows: the SQuAD protocol (span extraction, F1 and EM) is the estate's reader ruler, and the row note's "SQuAD human F1 86.8 per arxiv:1606.05250" is quoted from this abstract.
[devlin2018] Devlin, J., Chang, M.-W., Lee, K., Toutanova, K. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. NAACL 2019; arXiv:1810.04805. SQuAD v1.1 Test F1 93.2 (abstract). publisher · read in our library knowledge/fetched/cmp_deepresearch_devlin2018.html · pin hffa5d3591fa543ce7b833bea05f1455d40a6be8680417720f09da8807ada6abe · accessed 2026-08-18 · published-paperGrounds: The "Multi-hop QA reader" row note: "published SQuAD v1.1 SOTA = F1 932 (BERT, arxiv:1810.04805)" -- the 93.2 F1 is quoted from this abstract, and our reader's 510 is graded as 55 percent of it, honestly BEHIND.
[yang2018hotpot] Yang, Z., Qi, P., Zhang, S., Bengio, Y., Cohen, W.W., Salakhutdinov, R., Manning, C.D. HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering. EMNLP 2018; arXiv:1809.09600. publisher · read in our library knowledge/fetched/cmp_deepresearch_yang2018hotpot.html · pin h1d2aba7b7aa74242566af15165d3861311bc478db918ebf7a28ebc94d8b20ffd · accessed 2026-08-18 · datasetGrounds: The "Multi-hop QA reader" row: the Hotpot 505 figure in the note is measured on this dataset's multi-hop protocol; it is the second of the three benchmarks the reader ruler runs.
[trivedi2022musique] Trivedi, H., Balasubramanian, N., Khot, T., Sabharwal, A. MuSiQue: Multihop Questions via Single-hop Question Composition. TACL 2022; arXiv:2108.00573. publisher · read in our library knowledge/fetched/cmp_deepresearch_trivedi2022musique.html · pin hd8ca40a3303d9287245cad135d31a338a646f67e97764dd3ce4c59c321a578c7 · accessed 2026-08-18 · datasetGrounds: The "Multi-hop QA reader" row: "MuSiQue 122 ties the composition wall" -- MuSiQue is built by composing single-hop questions so that shortcut readers fail, which is the wall the note names; the third benchmark of the reader ruler.
[shao2024grpo] Shao, Z., Wang, P., Zhu, Q. et al. DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models. arXiv:2402.03300, 2024 -- introduces Group Relative Policy Optimization (GRPO), a PPO variant (abstract). publisher · read in our library knowledge/fetched/cmp_deepresearch_shao2024grpo.html · pin h32e37ca6d2af1522b1372c824d8ace3f400f44b3701a3cc69fd624254a39aeb5 · accessed 2026-08-18 · published-paperGrounds: The "RL-trained research agent" row note ("DeepResearcher / Tongyi GRPO-trained lead; Nishi none"): GRPO is defined in this paper, so the training method the row credits the leaders with has a single published definition, and the _ABSENT_ nx_research_agent watch names what a sovereign no-float RL loop must implement.
[arxiv-api] arXiv. arXiv API User's Manual (info.arxiv.org): the search_query grammar with field prefixes such as all: ti: au:, the Boolean operators AND OR ANDNOT, sortBy submittedDate, and the Atom feed schema returned by the query endpoint. publisher · read in our library knowledge/fetched/cmp_deepresearch_arxivapi.html · pin hf6c3ce2936f629e03fd42faf911604a7b0677ae3475c9c8608a543e9a7579671 · accessed 2026-08-20 · vendor-docGrounds: The "Capability-gap proposals diffed against our own board" row and knowledge/frontier_sources.conf: this manual is why every source row carries an explicit term-join operator. A space-joined search_query does NOT AND its terms, so an unjoined seed returns the newest submissions of the whole archive with HTTP 200 -- measured on the first metahuman replay, which proposed a Bose-Einstein condensate paper against a skin-shader row.
[openalex-api] OpenAlex. API Overview (docs.openalex.org): a free and open catalogue of scholarly works behind a public REST API that requires no key, with works filterable and sortable by publication date. publisher · read in our library knowledge/fetched/cmp_deepresearch_openalex.html · pin ha4405ec3a16a534736c39084c43a6c945b3850c27c080a0c4eed4798cd7ae0bf · accessed 2026-08-20 · vendor-docGrounds: The "Frontier noticing across every /compare domain on a beat" watch row: a fleet sweep is a rate-limit problem before it is a scheduling problem, and a second key-free scholarly index is what makes the wide sweep affordable without hammering a single source. It names the source row that rung must admit.
[arxiv_dr_202609] arXiv API query all:"deep research" AND cat:cs.AI, the 40 most recent by submission date, retrieved 2026-09-06. Enumerated in full: 41 of 41 title elements, 3 carrying a 2026-09 published date. September entries: 2609.02253 APEx (distillation of agent procedural experience for adaptive deep research question answering), 2609.01058 ARISE-RL (agentic rubric-grounded iterative self-evolution with reinforcement learning), 2609.00935 DualStake (dual-path confidence calibration in deep research agents). Adjacent August entries relied on: 2608.28643 Redesigning and Auditing Deep Research Writing for Faithful Reports, 2607.20891 Is Deep Research Reliable? Misleading Knowledge Induces False Conclusions, 2608.08389 Not Worth Another Token (marginal value estimation for efficient deep research agents). publisher · read in our library knowledge/fetched/cmp_deepresearch_arxiv_deepresearch_202609.atom · pin h31d7f18555f479d3227530e6851833e9d44b67d8e4d8e45de1bbfced51028fa5 · accessed 2026-09-06 · mirrored-indexGrounds: THE DATED BAR FOR THIS BOARD. Added 2026-09-06 by the researchreach lane because nx_compare_rank reported deepresearch BARFRESH verdict=NO-BAR and LADDER verdict=NO-LADDER with 13 rungs and 0 placed, which means its targets were undated and a seat could climb toward a dead bar with every rung honest. This listing is the evidence that the bar month is the CURRENT month and not an inherited one. STRICTLY ADDITIVE: no existing row on this board was edited.
[scoutforge-methodology-20260919] ScoutForge published scoring methodology; vendor claims, not independently executed audit. publisher · read in our library knowledge/fetched/rf_6d95efe0.raw · pin h7791603d1bcbd09822f39f08aa21991628786fe174ff180fdef49f8250156da5 · accessed 2026-09-19 · vendor-docGrounds: Six dimensions; AI editorial text; tooling evidence; undisclosed exact weighting. Benchmark presentation and disclosure, not score ground truth.
[scoutforge-remnus-20260919] ScoutForge Remnus worked review: observed internal inconsistencies. publisher · read in our library knowledge/fetched/rf_8ff8ac52.raw · pin h9f9367fe70665274e249e0a9e7235b76ea2994691f4b06c9c7517f26330974e5 · accessed 2026-09-19 · vendor-reviewGrounds: Page says 12 tools/4 resources/5 prompts versus 19/5/7; accessibility dimension70 versus prose55. Different Lighthouse81 distinguished from editorial score. Product claims not independently verified.

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