omh-tech-debt-audit
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from a repository (source code, manifests, and git logs), creating a surface for indirect prompt injection where malicious instructions embedded in the codebase could influence the agent's behavior during the audit.
- Ingestion points: The skill explicitly instructs the agent to read manifests, rank churn from git logs, and analyze source files across nine dimensions as detailed in
references/debt-dimensions.md. - Boundary markers: There are no explicit instructions or delimiters provided to help the agent distinguish between code/data being audited and instructions it should follow, which could lead to accidental obedience of instructions found in the code.
- Capability inventory: The skill utilizes shell command execution (intended for the operator's terminal) and file system access to gather evidence.
- Sanitization: The instructions do not mention any sanitization, escaping, or validation of the content read from the repository files before processing.
Audit Metadata