Repository navigation
fix: master CI red + live e2e run 8 (second intent) gate false positives - #12
Merged
Merged
Conversation
Without ctx-symbols, gate-scaffold-verify found a symbol's file with `(fn|...).*sym`, which also matched `fn sub_works` in tests/. Which file came first depended on directory order, so master CI failed two conformance cases that pass locally. It now uses the same name-anchored definition match as symbol_count. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015QxyVf3NgwsgRyYpBgDDTC
make e2e-live YES=1 INTENT=wordfreq. Checks that planning made both stories. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015QxyVf3NgwsgRyYpBgDDTC
…ged as duplicates Live e2e run 8 (wordfreq, two stories): 12/12 but 54 blocks. - path_matches_any split globs with pathname expansion on, so `tests/**` became the files that existed and an in-scope tests/common/mod.rs was blocked (29 GREEN blocks). - ctx-symbols conflicts flagged `mod common;` and local helpers in separate integration-test crates as HIGH duplicates (57 structural-review halts). The grep fallback skips tests/ too. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015QxyVf3NgwsgRyYpBgDDTC
Live e2e run 9 (wordfreq): 12/12, 29 blocks. - ensure-tools stamped its build with the plugin version only, so the ctx-symbols fix never reached the run. The stamp now includes a fingerprint of the tools' sources. scripts/test/ensure-tools.test.py (fake cargo). - The planner named story dirs S1-01/ and retried 56 times on a bare 'no story dir'. gate-stage2-complete now names the expected sN-NN-<slug>/ and the misnamed dir it found; the planner persona states the naming. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015QxyVf3NgwsgRyYpBgDDTC
adeelahmad
pushed a commit
that referenced
this pull request
Oct 18, 2026
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014j4wLin3oJTtnvjE9Lq2VV
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
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.
Requested by Adeel · project thread
Before:
After: all three causes are fixed, and
make e2e-live YES=1 INTENT=wordfreqis now a supported run.How:
(fn|…).*sym. That also matchedfn sub_worksin tests/, and the result depended on directory order. The fallback now uses the same name-anchored match assymbol_count.path_matches_any: it split scope globs with pathname expansion on, sotests/**turned into whatever files existed in the cwd. An in-scopetests/common/mod.rswas then blocked, 29 times in run 8. Globs are now read without expansion.ctx-symbols conflicts: it reportedmod common;and local helpers in separate integration-test files as HIGH duplicates, which halted the structural review 57 times. Eachtests/*.rsfile is its own crate, so a name that appears once per test file is no longer flagged. A test that redefines asrc/symbol is still HIGH. The grep fallback ingate-structural-integritynow skipstests/.Evidence:
dir/**scope is accepted. It failed first, then passed; the suite is 44/44.ctx-symbols conflictsreports 0 HIGH findings, down from 2 names repeated across every review.make ciis green locally.🤖 Generated with Claude Code
https://claude.ai/code/session_015QxyVf3NgwsgRyYpBgDDTC