[garrytan]

In the first full paid /cso runtime qualification (evaluator run 38995081113, gstack e23f7b8), repairs never ran. Across 42 comprehensive cells the agent recorded 148 repair reviews, but no report has a verification-attempt event, and heldOutRepairEachStack scored 0/24 on every runtime row. Two stacked helper bugs caused it.

Read-only snapshot files. preparePatchedSource (lib/cso/verification.ts) copies the snapshot, including its 0444 file modes, and then writes the patched file in place. For a non-root producer, which is how the evaluator and real users run, that write fails with EACCES. The CLI's catch-all then reports it as the generic "unexpected or unsafe input" error. The helper now makes an existing file owner-writable for the write and restores the original mode afterwards.

Runtime profile alias. runtime-plan returned the stack alias (node), and the skill tells the agent to copy that value into the verification request. Repair-bundle validation, however, requires the qualified runtime ID (node-24.4.0-amd64), so verify would still have failed once the first bug was fixed. Now runtime-plan returns the full ID whenever the runtime catalog resolves one. verify rejects any other profile with an INVALID_ARGUMENT that names the expected ID, and it does so before an attempt is counted.

Tests.

• test/cso-contracts.test.ts: a new non-root regression test patches a 0444 snapshot file and checks that the mode is kept. It fails with EACCES without the fix.
• test/cso-node-lifecycle-integration.test.ts: now asserts that runtime-plan returns cli.runtime.id, and builds the verification request from the plan's profile instead of hard-coding the runtime ID. That hard-coded ID is why the integration test never caught the alias bug.

End-to-end check. The fix was also verified on a real qualification cell, running as cso-producer with a unit built under the CI-pinned Bun 1.4.2. The finding ended at repair=runtime_tested and reproduction=reproduced.

Checks: bun run typecheck, and bun test test/cso-contracts.test.ts test/cso-cli.test.ts (81 pass). The Docker lifecycle suite needs a qualified node image and runs in the native validation lane.

Unstamped; the version slot will come from GStack Minder.

[trunk-io]

😎 This pull request was merged.

[github-actions]

E2E Evals: ✅ PASS

[test:paid] VERDICT GREEN — lane gate/pr, attempt 1
  rule 0/0 · behavior 0/0 · judge 0/0 · quarantined 0 (0 failing the lane)
  SKIPPED 0 · INFRA 0 · INCOMPLETE 0 · unattributed 0 · ACTION REQUIRED 0
  wall 0m00s · cost $0.00

0 executed, 0 reused rule/judge records | 0 manual accepted (unscored; no score-cache credit) | $0 rule/judge cost | reconcile exit: 0

Profile: pr / mode pr; selected behaviors: 0, judges: 0. Deferred to scheduled/release coverage: 3 behaviors and 0 changed prompt files. Deferred checks did not run and receive no PR-pass credit.

Rule and judge shards

Shard	Automated result	Manual/unscored	Executed	Reused	Status	Cost

Fail-closed reconciliation

[test:paid] report: 1/1 slices, 0 planned shards, tier=gate, attempt 1
[test:paid] coverage: profile=pr mode=pr; selected E2E=0, judges=0
[test:paid] deferred: 3 broad behaviors, 0 changed prompts without quick live coverage; these are not PR passes
[test:paid] collector final outcomes: 0 executed, 0 reused; 0 passed, 0 failed, 0 manual accepted (unscored; no score-cache credit) (0 attempt records from 0 collectors; every record counts)
[test:paid] VERDICT GREEN — lane gate/pr, attempt 1
  rule 0/0 · behavior 0/0 · judge 0/0 · quarantined 0 (0 failing the lane)
  SKIPPED 0 · INFRA 0 · INCOMPLETE 0 · unattributed 0 · ACTION REQUIRED 0
  wall 0m00s · cost $0.00
[test:paid] report: every planned shard accounted and passed

---

Sliced lane: planner → duration-packed executors → fail-closed report. Behavior cases run a pre-registered 3-trial panel (PASS at 2/3 with no contract violation); a PASS 2/3 is shown with its failed trial, never as a clean pass. Reused results retain their original provenance and expiry.