nightly: engine before the verbs + Goldpath.Cli mutation debt paid (62 % → 97 %) - #173
Merged
Conversation
…ation debt paid Two reds since 2026-08-06, both from the 2026-08-05 CLI PRs (#143–#147): 1. GmSixGateway: 'goldpath new service' gates itself through specdrift, but validate-gm.sh installed the engine only at spec-lint — AFTER the post-generation verbs, into a tool-path not on PATH. The engine is now resolved ONCE before any verb and handed to the CLI through its own GOLDPATH_SPECDRIFT hook, so 'new service', 'new gateway' and spec-lint all run the same 0.4.2 pin. Proven locally end to end (new service → new gateway → build → spec-lint → smoke green). 2. mutation (Goldpath.Cli): 62.35 % against a 70 % break. The four PRs added ~1,600 lines of command code with thin tests and the gate runs only nightly, so it was never seen on a PR. Seven new test files pin the generated files byte-exact, every refusal message and exit code, every engine/process invocation, and every branch the survivors lived on. Score now 96.95 % (killed 1650, survived 42 — the remaining ones are equivalent mutants, listed in the test files' comments where relevant). 133 → 368 tests, ~0.6 s. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Nightly has been red every night since 2026-08-06 on two jobs. Both trace to the 2026-08-05 CLI PRs (#143–#147).
GmSixGateway —
goldpath new servicegates its result through specdrift, butvalidate-gm.shinstalled the engine only at the spec-lint step (after the post-generation verbs, into a tool-path not on PATH) →could not run 'specdrift'. Fix: resolve the pinned engine once, BEFORE any verb, and hand it to the CLI via its existingGOLDPATH_SPECDRIFThook. Proven locally: new service → new gateway → build → spec-lint → smoke, green.mutation (Goldpath.Cli) — 62.35 % vs the 70 % break (528 survivors; the gate is nightly-only so the PRs never saw it). Seven new test files pin generated files byte-exact, every refusal message/exit code, every process invocation and branch. Score: 96.95 % (1650 killed / 42 survived — the remainder are equivalent mutants). Suite 133 → 368 tests, still ~0.6 s. No production code changed; the threshold was NOT touched.
After merge:
gh workflow run nightly.ymlto confirm before tonight's schedule.🤖 Generated with Claude Code