watch-review-requests
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external GitHub Pull Requests, specifically titles and metadata. This information is used to title new agent threads, which serves as a potential surface for indirect prompt injection if the external data contains instructions for the AI.
- [COMMAND_EXECUTION]: The Python script executes the
ghcommand-line tool via thesubprocessmodule. The implementation uses a list of arguments rather than a shell string, which is a secure practice that effectively mitigates shell injection risks.
Audit Metadata