test_bootstrap_audit: the SUITE-DEPENDS marker was stale — #20s gated caller now runs - #652
Conversation
…d caller now runs #20's close condition names this file by name, and its parenthetical was still true 19 days later: "CALLER THAT STILL RUNS IT (#381): tools/bootstrap-audit.py classifying a deliberately broken bootstrap and reporting the failure, IN A GATED SUITE. ⚠ Note test_bootstrap_audit currently declares `# SUITE-DEPENDS:` and is therefore gated nowhere (#286) — so the caller exists and does not run." ⇒ MEASURED BOTH DIRECTIONS, because "it passes" is not enough. A suite can pass by SKIPPING the legs that need the thing — which is exactly what the marker's own reason warns of ("a fake would make its control pass without the thing it controls for"): with a live registry rc 0 · 12 check lines env -i, HOME=/nonexistent rc 0 · 12 check lines · output BYTE-IDENTICAL ⛔ CONTROL, same bare env test_stranded_branches.py -> rc 1, FileNotFoundError: 'gh' (a real dependency) ⇒ No leg reads the registry. The declaration excluded the suite from the gate for a dependency it does not have. AT THE GATE, and the number MOVES with the change — which is what makes it a measurement: before ran 57 hermetic suite(s) · 57 passed · skipped 4 after ran 58 hermetic suite(s) · 58 passed · skipped 3 · 0 FINDINGS ★ I SWEPT THE OTHER THREE with the same probe, and only this one was stale: test_runmarker bare rc 1, output DIFFERS genuine (injects into pane-binding, which exits 2 on an empty registry) test_stranded_branches bare rc 1, output DIFFERS genuine (gh, network + auth) test_grant_check bare rc 0, IDENTICAL ⚠ NOT stale — its marker states a deliberate judgement: "MEASURED PASSING on a runner (run 32345860216); listed here only because a network-dependent gate is a flake source." That is an author's call about flakiness, not a claim I can refute by running it once. Left alone. ⚠ This ADDS a suite to the gating population. It passes hermetically today; if it ever fails there, that is a finding rather than a regression in this change. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
🛡️ Sentinel PR review1 file(s) changed · 0 introduced by this diff (secrets+SAST) · dependencies unchanged — SCA/CVE not re-scanned. Advisory — the fail-closed gate is the post-merge pentest. Findings — ranked by criticalityNo issues found on the changed surface. 🤖 Code review (Flynn)No issues found. Scan summary
|
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Advanced Run ID: 📒 Files selected for processing (1)
Included review availability: Your plan provides up to 1 included review per hour; 0 remain after this review. 📝 WalkthroughWalkthroughThe change removes an obsolete ChangesBootstrap audit metadata
Estimated code review effort: 1 (Trivial) | ~2 minutes Merge Risk: ⚪ Minimal · up to This removes a stale suite dependency declaration so the bootstrap audit suite is included in hermetic gating. No remaining merge-blocking risk is identified. 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Addresses #20's "caller that still runs it" leg. TEAMLEAD (session
15b69750), 2026-09-08.The claim #20 made, still true 19 days later
Measured on
origin/main: 4 of 61 suites declare the marker, and this is one.⇒ The marker was stale, and "it passes" was not enough to show that
The declaration's own reason warns of the trap: "a fake would make its control pass without the thing it controls for." A suite can pass by skipping the legs that need the thing. So both directions, and a control:
No leg reads the registry. The declaration excluded the suite from the gate for a dependency it does not have.
At the gate — and the number moves
⇒ That movement is what makes it a measurement. Earlier today I ran a marker test where
skipped 4came back identical with and without the change — a number that does not move is not evidence, and I nearly recorded it as one.★ I swept the other three, and only this one was stale
test_runmarkerpane-binding, which exits 2 on an empty registry before reaching the injectiontest_stranded_branchesgh, network + authtest_grant_check⚠ The last row is the one worth pausing on: the same measurement (
bare rc 0, identical output) means "stale" in one file and "deliberate" in another, and only the stated reason distinguishes them. A sweep that acted on the measurement alone would have removed both.What this does not claim
⚠ It adds a suite to the gating population. It passes hermetically today; if it ever fails there, that is a finding — the whole point of gating it — rather than a regression in this change.
⚠ And it does not close #20, whose predicate is about a pane reporting bootstrap failure in its ROLE-READY line. This lands one named leg of its condition.
🤖 Generated with Claude Code
Summary by CodeRabbit