check-impl-against-spec

Pass

Audited by Gen Agent Trust Hub on Nov 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it ingests and processes external, untrusted content to influence its logic.
  • Ingestion points: The skill reads spec_context.md, pr_diff.txt, and pr_description.txt directly from the working tree, which are often controlled by external PR authors.
  • Boundary markers: The instructions lack explicit delimiters or "ignore embedded instructions" warnings for the agent when processing these external inputs.
  • Capability inventory: The skill reads file contents and generates findings for review.json. While it does not perform network operations or direct shell execution, the review outcome can be manipulated by instructions hidden within the spec or diff files.
  • Sanitization: No sanitization or validation steps are defined for the content extracted from the PR files before analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 19, 2026, 01:12 AM