nx_conn_catalog.nx
buildroot/runtime/nx_conn_catalog.nx
about
nx_conn_catalog.nx -- the CONNECTION / TRANSPORT / CRYPTO capability catalogue.
Operator 2026-07-17: "our catalogue is shitty for exposing these to you and thats why they need to
become standardized apis and microservices and all that catalogued and exposed on a state of the art
level." The old nx_capability_catalog tracks 16 hand-typed needs and OMITS the entire networking stack
(QUIC, HTTP/2, HTTP/3, TLS, PQ, WebRTC) -- so the richest capabilities are BUILT but INVISIBLE to agents.
This catalogue makes the real stack DISCOVERABLE + measures the gap that matters: BUILT-BUT-UNEXPOSED
(a capability that exists on disk but has NO standardized MCP/API verb = an agent cannot find or call it).
For each capability: BUILT? (organ opens on disk) and EXPOSED? (a standardized MCP/API verb per the
exposure policy, knowledge/mcp/coverage_map.txt). The exposure debt is the ranked roadmap: wrap each
BUILT-BUT-UNEXPOSED capability as ONE well-typed verb (never blind-wrap primitives -- exposure_policy).
Emits a durable, discoverable catalogue (knowledge/conn_catalog.md) + a log line. license_tier: ORIGINAL
dependencies 1 imports · 0 importers
imports: nx_gate.nx
imported by: nobody (leaf or entry point)
call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown
structs
| none |
consts
| none |
functions
| 16 | func cc_have(path: *u8) -> i64 { let fd: i64 = sys_openat_rd(path); if fd < 0 { return 0 } sys_close(fd); return 1 } called by 1: cc_cap |
| 17 | func cw(fd: i64, s: *u8) -> i64 { var n: i64=0; while s[n]!=(0 as u8){n=n+1} sys_write(fd,s,n); return 0 } |
| 18 | func cn(fd: i64, v: i64) -> i64 called by 1: main |
| 28 | func cc_cap(cat: *u8, name: *u8, path: *u8, exposed: i64, note: *u8, mdfd: i64, built: *i64, exp: *i64, unexp: *i64, tot: *i64) -> i64 |
| 50 | func main() -> i64 |