complexity-review
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which can serve as a vector for indirect prompt injection attacks.
- Ingestion points: The skill is designed to analyze code diffs (
git diff), pull requests, modules, and design documents provided by the user or environment. - Boundary markers: The instructions do not define clear delimiters or specific warnings to the agent to treat the code/design input strictly as data, which could allow instructions hidden in comments or documentation to be executed by the agent.
- Capability inventory: The skill primarily performs text analysis and reporting, though it functions within an agent environment that has access to local file systems and version control tools.
- Sanitization: No sanitization, escaping, or filtering of the ingested code content is specified.
Audit Metadata