Skip to content

[compiler-threat-spec] docs: sync CTR-025 framework self-prompt misattribution spec - #53037

Merged
pelikhan merged 2 commits into
mainfrom
copilot/sync-ctr-025-framework-spec
Aug 16, 2026
Merged

[compiler-threat-spec] docs: sync CTR-025 framework self-prompt misattribution spec#53037
pelikhan merged 2 commits into
mainfrom
copilot/sync-ctr-025-framework-spec

Conversation

Copilot AI commented Aug 16, 2026

Copy link
Copy Markdown
Contributor

This PR closes a spec/implementation drift: CTR-025 behavior (framework <system> block stripping in threat-detection setup) was implemented and tested but not documented in the threat spec/compliance mapping.
It updates the threat-detection specification and conformance map so the normative catalog, implementation mapping, and test IDs align.

  • Spec sync (v1.0.23)

    • Bumped specs/compiler-threat-detection-spec.md to 1.0.23.
    • Added Section 2 sync-table entry (1.0.23 → minimum binary v0.87.1, runtime-only behavior, no .lock.yml schema change).
    • Added changelog entry for 1.0.23 (2026-08-16).
  • Rule + mapping coverage

    • Added CTR-025 Framework Self-Prompt Misattribution to Section 5.1.
    • Added Section 7.1 implementation mapping to:
      • actions/setup/js/setup_threat_detection.cjs (stripFrameworkSystemBlock, SYSTEM_BLOCK_REMOVED_MARKER)
      • actions/setup/js/setup_threat_detection.test.cjs
  • Conformance ID alignment

    • Added T-CTR-039 in Section 8.1 for CTR-025.
    • Updated specs/compiler-threat-detection-compliance/README.md with CTR-025 → T-CTR-039.
  • Release note entry

    • Added .changeset/sync-ctr-025-system-block-spec.md (patch) to record the documentation sync.
// Documented behavior for CTR-025:
const strippedPrompt = stripFrameworkSystemBlock(rawPrompt);
// strips only leading framework <system>...</system> block
// preserves later <system> lookalikes for threat analysis

Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>
Copilot AI changed the title [WIP] Sync CTR-025 framework self-prompt misattribution spec [compiler-threat-spec] docs: sync CTR-025 framework self-prompt misattribution spec Aug 16, 2026
Copilot AI requested a review from pelikhan August 16, 2026 04:31
@pelikhan
pelikhan marked this pull request as ready for review August 16, 2026 04:33
Copilot AI balanced review requested due to automatic review settings August 16, 2026 04:33
@pelikhan
pelikhan merged commit b5d5f18 into main Aug 16, 2026
@pelikhan
pelikhan deleted the copilot/sync-ctr-025-framework-spec branch August 16, 2026 04:33

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Documents the existing CTR-025 framework prompt-stripping behavior and aligns its conformance mapping.

Changes:

  • Adds CTR-025 and T-CTR-039 to specification version 1.0.23.
  • Maps the rule to its implementation and tests.
  • Adds compliance and release-note entries.
Show a summary per file
File Description
specs/compiler-threat-detection-spec.md Adds the normative rule, mappings, test ID, and changelog.
specs/compiler-threat-detection-compliance/README.md Maps CTR-025 to T-CTR-039.
.changeset/sync-ctr-025-system-block-spec.md Records the documentation patch.

Review details

💡 Add a code-review agent skill for context-aware, tailored reviews. Learn more in the docs.

Suppressed comments (1)

specs/compiler-threat-detection-spec.md:423

  • The changelog repeats the incorrect v0.87.1 minimum. Since the CTR-025 implementation is already present in v0.87.0, keep this entry aligned with the corrected sync-table version.
- Updated Section 2 spec-to-implementation sync table with version 1.0.23 mapped to minimum binary `v0.87.1` and no `.lock.yml` schema changes.
  • Files reviewed: 3/3 changed files
  • Comments generated: 1
  • Review effort level: Balanced


| Spec version | Minimum gh-aw binary version | Lock-file compatibility notes |
|--------------|------------------------------|-------------------------------|
| `1.0.23` | `v0.87.1` (or newer) | Adds normative coverage for framework self-prompt misattribution handling (CTR-025) in threat-detection setup by stripping only the leading framework-generated `<system>...</system>` block before analysis; no `.lock.yml` schema changes (runtime-only detection setup behavior). |
@github-actions

Copy link
Copy Markdown
Contributor

🎉 This pull request is included in a new release.

Release: v0.87.1

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[compiler-threat-spec] docs: sync CTR-025 framework self-prompt misattribution spec

3 participants