feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) - #1257
Conversation
|
Warning You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again! |
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
|
Warning Review limit reached
Next review available in: 32 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Organization UI Review profile: ASSERTIVE Plan: Pro Run ID: 📒 Files selected for processing (2)
📝 WalkthroughWalkthroughAdds a Bash guard that validates reusable-workflow caller inputs against declarations at pinned refs, wires it into lint CI, documents the contract, and adds fixtures and Bats tests for valid, invalid, regression, CLI, and unresolved-reference cases. ChangesReusable caller-input validation
Estimated code review effort: 4 (Complex) | ~45 minutes Sequence Diagram(s)sequenceDiagram
participant Lint
participant Validator
participant ReusableWorkflow
participant Bats
Lint->>Validator: run validate-caller-inputs.sh
Validator->>ReusableWorkflow: resolve workflow at pinned ref
ReusableWorkflow-->>Validator: return workflow_call inputs
Validator-->>Lint: pass, fail, or warning
Bats->>Validator: execute helper, pair, and scan tests
Validator-->>Bats: validation results
Possibly related issues
Suggested labels: 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 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 |
Dev-Lead — review-changes (no-changes)No changes were needed for this PR. |
CI Failure: SonarCloud Code AnalysisStep: Quality Gate — "C Security Rating on New Code" (required ≥ A) SonarCloud flagged 3 "Use full commit SHA hash for this dependency" security findings in the new test fixture workflows added by this PR: Suggested fix: In the three fixture |
Dev-Lead Fix CI — appliedPR: #1257 | SHA: |
Dev-Lead — review-changes (no-changes)No changes were needed for this PR. |
Dev-Lead — fix-bot-comment (no-changes)Agent reasoning |
|
Advisory bots were rate-limited; auto-approval is withheld until they recover. pr-review-sweep will re-review this PR after 2026-07-15T04:36:12Z. |
There was a problem hiding this comment.
Pull request overview
This PR adds a new CI guard to prevent “channel-skew” breakages where a caller workflow forwards with: inputs that are not declared (or misses required inputs) in the reusable workflow at the pinned @ref (issue #1253 / regression class #1034). It does so by introducing a validation script, wiring it into lint.yml, and adding unit/fixture tests that reproduce the failure mode.
Changes:
- Add
scripts/validate-caller-inputs.shto resolve pinned reusable refs and validate(a) forwarded ⊆ declared inputsand(b) required inputs forwarded. - Wire the new validation as a
validate-caller-inputsjob in.github/workflows/lint.yml. - Add Bats unit tests plus fixture caller/reusable trees (including a #1034 regression fixture) and document the guard in
AGENTS.md.
Reviewed changes
Copilot reviewed 9 out of 9 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
scripts/validate-caller-inputs.sh |
Implements the reusable-at-pinned-ref resolution + forwarded-input contract validation. |
.github/workflows/lint.yml |
Adds the validate-caller-inputs CI job to run the new guard script. |
tests/dev-lead/unit/test_validate_caller_inputs.bats |
Unit + fixture-driven tests covering parsing, validation logic, and the #1034 regression. |
tests/dev-lead/fixtures/caller-inputs/reusable/dev-lead-reusable.yml |
Fixture reusable declaring optional + required inputs (and required secret) for validation tests. |
tests/dev-lead/fixtures/caller-inputs/good/.github/workflows/caller.yml |
Passing fixture caller that forwards only declared inputs and includes required ones. |
tests/dev-lead/fixtures/caller-inputs/regression-1034/.github/workflows/caller.yml |
Regression fixture that forwards an undeclared input and must fail. |
tests/dev-lead/fixtures/caller-inputs/missing-required/.github/workflows/caller.yml |
Negative fixture omitting a required input and must fail. |
AGENTS.md |
Documents the new guard/job and its rationale/behavior. |
Superseded by automated re-review at
|
…mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Dev-Lead — waiting on PR blockers (intent: review-changes)PR: #1257 |
|
Note @don-petry I reviewed this PR and no code changes were needed, but it still has blocking checks or reviews (failing or cancelled checks, or changes-requested reviews), so I cannot mark it done yet. I'll re-check automatically. |
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In @.github/workflows/lint.yml:
- Around line 144-151: Document the intentional default credential persistence
on the actions/checkout step in the workflow, noting that
scripts/validate-caller-inputs.sh and vci_resolve_reusable require persisted
credentials for authenticated git fetch origin resolution of same-repository
channel tags; do not set persist-credentials to false, and optionally scope the
job permissions to the minimum read-only access required.
In `@scripts/validate-caller-inputs.sh`:
- Around line 63-73: Update the reusable-workflow reference matching in the
relevant validation function to recognize both .yml and .yaml paths, while
preserving the existing extraction of repo_slug, wf_path, and ref. Also update
the workflow scan around the existing .yml glob to process .yaml files, ensuring
references using either extension are validated and unresolved cases still emit
the existing warning.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro
Run ID: 128a6000-2558-4015-8991-fd91d8be758c
📒 Files selected for processing (9)
.github/workflows/lint.ymlAGENTS.mdscripts/validate-caller-inputs.shsonar-project.propertiestests/dev-lead/fixtures/caller-inputs/good/.github/workflows/caller.ymltests/dev-lead/fixtures/caller-inputs/missing-required/.github/workflows/caller.ymltests/dev-lead/fixtures/caller-inputs/regression-1034/.github/workflows/caller.ymltests/dev-lead/fixtures/caller-inputs/reusable/dev-lead-reusable.ymltests/dev-lead/unit/test_validate_caller_inputs.bats
|
donpetry-bot
left a comment
donpetry-bot
left a comment
There was a problem hiding this comment.
Automated review — APPROVED ✓
Risk: MEDIUM
Reviewed commit: 077c48b1683a8368246e648260a06070b3029352
Review mode: triage-approved (single reviewer)
Summary
Adds the #1052 Part A CI guard (issue #1253): a new validate-caller-inputs job in lint.yml backed by scripts/validate-caller-inputs.sh, which resolves every reusable-workflow caller's pinned @ref and asserts forwarded with: inputs are declared there and required inputs are forwarded. Includes bats unit + fixture tests (with a #1034 regression fixture), AGENTS.md documentation, and a Sonar exclusion for the new fixtures. Triage cleared this as low-risk; confirmation review agrees the implementation is sound — classified MEDIUM here only because it adds non-trivial CI logic.
Linked issue analysis
Closes #1253 ([#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref). All acceptance criteria are met: (1) guard FAILS on the #1034 pattern via the regression-1034 fixture and the new lint job passes on this branch; (2) shellcheck is green and all existing suites stay green; (3) the change is additive, the checkout action is SHA-pinned (actions/checkout v6.0.3), and review behaviour is unchanged. Unresolvable refs soft-pass with a logged ::warning:: as specified, never silently.
Findings
No blocking findings.
- Security: git fetch uses '--' before refs (option-injection guard, addressing Copilot's earlier findings — threads resolved); cross-repo resolution is opt-in via VCI_RESOLVE_CROSS_REPO; fetched workflow content is parsed as text, never executed. persist-credentials is intentionally left true with a documented rationale (required for same-repo tag resolution; setting false would silently route every caller into the soft-pass path).
- Review threads: all 5 threads (3 Copilot, 2 CodeRabbit) are resolved; CodeRabbit's changes-requested review was addressed and dismissed, and CodeRabbit subsequently approved.
- Secret scan: run_secret_scanning MCP tool not available in this environment; gitleaks CI check is green and the diff introduces no credential-like content.
- Since the prior review SHA (9f6b688), the only change is a merge of main into the branch — no new PR-specific code.
CI status
All required checks green: Lint (incl. the new validate-caller-inputs job), shellcheck, bats, unit tests, CodeQL (actions + python), Secret scan (gitleaks), SonarCloud quality gate, AgentShield, Agent Security Scan, holdout-guard, template-drift, and permission/structure guards. Skipped checks are conditional ecosystem audits (npm/pip/cargo/go audits, dependabot-automerge, ci-relay) — not applicable to this PR.
Reviewed automatically by the PR-review agent (single-reviewer mode: fable 5). Reply if you need a human review.
Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run).
…the fleet-monitor workflow behind a human go/no-go (#1275) * feat: implement issue #1152 — [Phase 4] Wire opt-in remediation into the fleet-monitor workflow behind a human go/no-go * fix(reviews): address review comments [skip ci-relay] * style(fleet-gate): scope strict mode to direct-exec (no source leak) Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run). --------- Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com>
…puts ⊆ declared inputs at the pinned @ref (+ #1034 regression) (#1257) * feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) * fix(ci): auto-fix for SonarCloud Code Analysis [skip ci-relay] * fix(vci): harden git fetch against argument injection + scope strict mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit. * fix(reviews): address review comments [skip ci-relay] --------- Co-authored-by: donpetry-bot <{}+donpetry-bot@users.noreply.github.com> Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com> Co-authored-by: Don Petry Bot <donpetry+bot@gmail.com>
…puts ⊆ declared inputs at the pinned @ref (+ #1034 regression) (#1257) * feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) * fix(ci): auto-fix for SonarCloud Code Analysis [skip ci-relay] * fix(vci): harden git fetch against argument injection + scope strict mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit. * fix(reviews): address review comments [skip ci-relay] --------- Co-authored-by: donpetry-bot <{}+donpetry-bot@users.noreply.github.com> Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com> Co-authored-by: Don Petry Bot <donpetry+bot@gmail.com>
…the fleet-monitor workflow behind a human go/no-go (#1275) * feat: implement issue #1152 — [Phase 4] Wire opt-in remediation into the fleet-monitor workflow behind a human go/no-go * fix(reviews): address review comments [skip ci-relay] * style(fleet-gate): scope strict mode to direct-exec (no source leak) Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run). --------- Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com>
…puts ⊆ declared inputs at the pinned @ref (+ #1034 regression) (#1257) * feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) * fix(ci): auto-fix for SonarCloud Code Analysis [skip ci-relay] * fix(vci): harden git fetch against argument injection + scope strict mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit. * fix(reviews): address review comments [skip ci-relay] --------- Co-authored-by: donpetry-bot <{}+donpetry-bot@users.noreply.github.com> Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com> Co-authored-by: Don Petry Bot <donpetry+bot@gmail.com>
…the fleet-monitor workflow behind a human go/no-go (#1275) * feat: implement issue #1152 — [Phase 4] Wire opt-in remediation into the fleet-monitor workflow behind a human go/no-go * fix(reviews): address review comments [skip ci-relay] * style(fleet-gate): scope strict mode to direct-exec (no source leak) Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run). --------- Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com>
…puts ⊆ declared inputs at the pinned @ref (+ #1034 regression) (#1257) * feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) * fix(ci): auto-fix for SonarCloud Code Analysis [skip ci-relay] * fix(vci): harden git fetch against argument injection + scope strict mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit. * fix(reviews): address review comments [skip ci-relay] --------- Co-authored-by: donpetry-bot <{}+donpetry-bot@users.noreply.github.com> Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com> Co-authored-by: Don Petry Bot <donpetry+bot@gmail.com>
…the fleet-monitor workflow behind a human go/no-go (#1275) * feat: implement issue #1152 — [Phase 4] Wire opt-in remediation into the fleet-monitor workflow behind a human go/no-go * fix(reviews): address review comments [skip ci-relay] * style(fleet-gate): scope strict mode to direct-exec (no source leak) Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run). --------- Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com>
…puts ⊆ declared inputs at the pinned @ref (+ #1034 regression) (#1257) * feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) * fix(ci): auto-fix for SonarCloud Code Analysis [skip ci-relay] * fix(vci): harden git fetch against argument injection + scope strict mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit. * fix(reviews): address review comments [skip ci-relay] --------- Co-authored-by: donpetry-bot <{}+donpetry-bot@users.noreply.github.com> Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com> Co-authored-by: Don Petry Bot <donpetry+bot@gmail.com>
…the fleet-monitor workflow behind a human go/no-go (#1275) * feat: implement issue #1152 — [Phase 4] Wire opt-in remediation into the fleet-monitor workflow behind a human go/no-go * fix(reviews): address review comments [skip ci-relay] * style(fleet-gate): scope strict mode to direct-exec (no source leak) Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run). --------- Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com>
…puts ⊆ declared inputs at the pinned @ref (+ #1034 regression) (#1257) * feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) * fix(ci): auto-fix for SonarCloud Code Analysis [skip ci-relay] * fix(vci): harden git fetch against argument injection + scope strict mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit. * fix(reviews): address review comments [skip ci-relay] --------- Co-authored-by: donpetry-bot <{}+donpetry-bot@users.noreply.github.com> Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com> Co-authored-by: Don Petry Bot <donpetry+bot@gmail.com>
…the fleet-monitor workflow behind a human go/no-go (#1275) * feat: implement issue #1152 — [Phase 4] Wire opt-in remediation into the fleet-monitor workflow behind a human go/no-go * fix(reviews): address review comments [skip ci-relay] * style(fleet-gate): scope strict mode to direct-exec (no source leak) Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run). --------- Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com>
…puts ⊆ declared inputs at the pinned @ref (+ #1034 regression) (#1257) * feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) * fix(ci): auto-fix for SonarCloud Code Analysis [skip ci-relay] * fix(vci): harden git fetch against argument injection + scope strict mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit. * fix(reviews): address review comments [skip ci-relay] --------- Co-authored-by: donpetry-bot <{}+donpetry-bot@users.noreply.github.com> Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com> Co-authored-by: Don Petry Bot <donpetry+bot@gmail.com>
…the fleet-monitor workflow behind a human go/no-go (#1275) * feat: implement issue #1152 — [Phase 4] Wire opt-in remediation into the fleet-monitor workflow behind a human go/no-go * fix(reviews): address review comments [skip ci-relay] * style(fleet-gate): scope strict mode to direct-exec (no source leak) Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run). --------- Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com>
…puts ⊆ declared inputs at the pinned @ref (+ #1034 regression) (#1257) * feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) * fix(ci): auto-fix for SonarCloud Code Analysis [skip ci-relay] * fix(vci): harden git fetch against argument injection + scope strict mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit. * fix(reviews): address review comments [skip ci-relay] --------- Co-authored-by: donpetry-bot <{}+donpetry-bot@users.noreply.github.com> Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com> Co-authored-by: Don Petry Bot <donpetry+bot@gmail.com>
…the fleet-monitor workflow behind a human go/no-go (#1275) * feat: implement issue #1152 — [Phase 4] Wire opt-in remediation into the fleet-monitor workflow behind a human go/no-go * fix(reviews): address review comments [skip ci-relay] * style(fleet-gate): scope strict mode to direct-exec (no source leak) Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run). --------- Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com>
…puts ⊆ declared inputs at the pinned @ref (+ #1034 regression) (#1257) * feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) * fix(ci): auto-fix for SonarCloud Code Analysis [skip ci-relay] * fix(vci): harden git fetch against argument injection + scope strict mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit. * fix(reviews): address review comments [skip ci-relay] --------- Co-authored-by: donpetry-bot <{}+donpetry-bot@users.noreply.github.com> Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com> Co-authored-by: Don Petry Bot <donpetry+bot@gmail.com>
…the fleet-monitor workflow behind a human go/no-go (#1275) * feat: implement issue #1152 — [Phase 4] Wire opt-in remediation into the fleet-monitor workflow behind a human go/no-go * fix(reviews): address review comments [skip ci-relay] * style(fleet-gate): scope strict mode to direct-exec (no source leak) Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run). --------- Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com>
…puts ⊆ declared inputs at the pinned @ref (+ #1034 regression) (#1257) * feat: implement issue #1253 — [#1052 A] CI guard: forwarded caller inputs ⊆ declared inputs at the pinned @ref (+ #1034 regression) * fix(ci): auto-fix for SonarCloud Code Analysis [skip ci-relay] * fix(vci): harden git fetch against argument injection + scope strict mode to direct-exec Addresses the pr-review/Copilot findings on #1257 (Part A of #1052): - Add end-of-options `--` before "$ref" in both git fetch calls so a ref beginning with `-` (from a PR-modifiable, CI-executed workflow file) is treated as a refspec, not parsed as a flag/option (verified: without `--` git errors "unknown option"; with `--` it treats it as a ref). - Move `set -euo pipefail` out of the top level into the execute-directly guard so sourcing the script (bats) no longer leaks strict mode into the caller's shell. shellcheck clean; all 16 bats tests pass; sourced no longer sets errexit. * fix(reviews): address review comments [skip ci-relay] --------- Co-authored-by: donpetry-bot <{}+donpetry-bot@users.noreply.github.com> Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com> Co-authored-by: Don Petry Bot <donpetry+bot@gmail.com>
…the fleet-monitor workflow behind a human go/no-go (#1275) * feat: implement issue #1152 — [Phase 4] Wire opt-in remediation into the fleet-monitor workflow behind a human go/no-go * fix(reviews): address review comments [skip ci-relay] * style(fleet-gate): scope strict mode to direct-exec (no source leak) Human-review polish on #1275: move `set -euo pipefail` out of the top level into the execute-directly guard, so sourcing fleet_remediate_gate.sh for the bats tests doesn't leak strict mode into the caller's shell — matches the repo standard codified in #1257/AGENTS.md. Pure/behaviour-preserving: shellcheck clean, all 10 gate bats pass, sourcing leaves errexit off, and the gate still fail-closes (schedule->dry-run, live dispatch w/ all preconditions->live, missing token/pilot->dry-run). --------- Co-authored-by: donpetry-bot <281750570+donpetry-bot@users.noreply.github.com>
Closes #1253
Implemented by dev-lead agent. Please review.
Summary by CodeRabbit
New Features
Documentation
Tests
Chores