agent-pr-creator
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill inherently processes external data that could contain malicious instructions or unexpected content designed to influence the agent's logic.
- Ingestion points: Reads external content during Phase 2 and Phase 3 via
git log,git diff, and repository-suppliedPULL_REQUEST_TEMPLATE.mdfiles. - Boundary markers: Does not specify clear delimiters or boundary instructions to isolate the parsed git content from the agent's core instructions.
- Capability inventory: Employs the
Bashtool to run critical commands such asgit pushandgh pr createusing parameters derived from the ingested data. - Sanitization: Lacks sanitization or validation routines to verify that commit messages or diff inputs do not introduce prompt injection vectors or malformed shell arguments.
Audit Metadata