golem15/msd-core
4
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs(1259-01): wire verify-phase consumer + ADR-550/FEATURES/reference + changeset

Browse Source 
- verify-phase.md: replace test-tier 'fail-closed/deferred' bullet with the check prohibition-enforcement
  enforcement step (locate -> fail-first -> run -> evidence -> green-or-hard-gate); update determine_status tree
- ADR-550 addendum: mark D5d enforcement half LANDED (#1259); cite ADR-857 open-question §147 + D6 (core verify rail)
- FEATURES §146: enforcement wording + add REQ-PROHIB-07; keep REQ-PROHIB-06 intact
- references/prohibition-probe.md: test-tier enforced + hard-gates via check prohibition-enforcement
- changeset (type: Changed) with the D5 '2 no-source-grep invalid cases, not 96' correction
This commit is contained in:
Dave
2026-06-15 12:47:00 -04:00
parent 8f428bdd41
commit ce01e1376b
5 changed files with 36 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
.changeset/1259-test-tier-enforcement.md Normal file
View File

@@ -0,0 +1,8 @@
---
type: Changed
pr: 1259
---
**Test-tier prohibitions are now a real, provable gate instead of a permanent, unsatisfiable `gaps_found`** — the deferred ENFORCEMENT half of ADR-550 Decision 5d (the "heavy half" that #644 / PR #1149 deferred) has landed. A new deterministic `check prohibition-enforcement` sub-command (authored as `src/prohibition-enforcement.cts`, compiled by `build:lib` to the gitignored `gsd-core/bin/lib/prohibition-enforcement.cjs`) is the missing PRODUCER: it locates the wired mechanical check, confirms it is fail-first (`regression-must-fail-first`), runs it, builds `enforcementEvidence`, and emits the `dispositionForProhibition()` verdict. The previously-unreachable green branch in `dispositionForProhibition()` is now reachable from the live pipeline — a test-tier prohibition with a PASSING wired check disposes `green` and can reach `passed`, while a missing, failing, or non-fail-first check hard-gates (flagged, never green → `gaps_found`) in BOTH interactive and autonomous modes (ADR-550 D4 / D3). `verify-phase.md` wires the consumer; the green/fail-closed policy in `src/probe-core.cts` is untouched. Both wired-check kinds are accepted (ADR-550 D2): a `node --test` negative test AND a lint/AST rule, anchored on the in-tree `no-source-grep` rule (dogfooding, ADR-550 D4). This enforcement seam is the concrete instance of ADR-857 open-question §147 and lands on the core verify rail, never in `capabilities/` (D6). (#1259)
**Correction to the issue body (#1259):** the issue's "96 invalid/error negative-proof cases" figure is wrong. The definitive count is **26 invalid cases / 23 error-expectation objects** across all rule-test files; the `no-source-grep` anchor itself has exactly **2 invalid cases** (`tests/eslint-rules.test.cjs`, the `.includes()` and `.match()` invalid blocks). Those 2 cases ARE genuine `regression-must-fail-first` proofs, so the anchor argument is unaffected — but the count is "the 2 `no-source-grep` invalid cases," not 96.