github-pr-query
Pass
Audited by Gen Agent Trust Hub on Apr 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the GitHub CLI (
gh) andjqto fetch and filter pull request data. The commands are constructed using variables that are properly handled within the bash script, following standard practices for CLI wrappers. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it processes untrusted data (pull request metadata) from external GitHub repositories.
- Ingestion points: External data is ingested in
query-prs.shvia thegh pr listcommand. - Boundary markers: The script does not implement specific delimiters or 'ignore' instructions to wrap the external data in its output.
- Capability inventory: The skill possesses capabilities to read repository data via
ghand process it locally usingjq. - Sanitization: There is no sanitization of the content (like PR titles or labels) retrieved from GitHub before it is returned to the agent context.
Audit Metadata