production-readiness-review
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze untrusted backend source code, creating a surface for indirect prompt injection attacks.
- Ingestion points: Source code files and Git diffs identified in the Scope Resolution section of SKILL.md.
- Boundary markers: The protocol includes mandatory "Why-Review" validation gates to stress-test findings and requires explicit file:line evidence for all claims to prevent obedience to embedded instructions.
- Capability inventory: The skill can spawn sub-agents (spawn_agent), execute local Python and Node.js scripts for graph analysis and signal scanning (.claude/scripts/), and write report files to disk.
- Sanitization: The skill employs an adversarial validation protocol to verify findings against actual code before any fixes are applied, reducing the likelihood of malicious code influencing the agent's output.
Audit Metadata