sec-diff-scan
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to run standard CLI tools for scope resolution and analysis. Evidence includes usage of
git diff,git diff --staged, and the GitHub CLI (gh). It also references local scripts such assecfind.pyfor ID generation and validation. - [INDIRECT_PROMPT_INJECTION]: The skill operates on untrusted code diffs and Pull Request metadata which constitutes a surface for indirect prompt injection attacks.
- Ingestion points: Results from
git diffandghcommands representing external/untrusted code changes. - Boundary markers: The instructions do not specify the use of delimiters or specific safety markers for processing the untrusted input content.
- Capability inventory: The agent possesses the capability to read local files and write results to the file system (findings.json).
- Sanitization: The procedure explicitly includes steps to mask secrets and tokens to prevent exposure.
Audit Metadata