pr
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes git and gh commands to automate the pull request workflow. These operations depend on user input and the repository's state.\n- [EXTERNAL_DOWNLOADS]: Instructions reference the use of third-party GitHub CLI extensions (gh-image, gh-attach). While installation requires user confirmation, these are external dependencies not managed by the skill itself.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository data to build PR summaries and bodies. Instructions in source diffs or template files could potentially manipulate the agent's actions.\n
- Ingestion points: Workflow steps 5 and 6 in SKILL.md (diff inspection and template reading).\n
- Boundary markers: Absent from the instructions.\n
- Capability inventory: Shell command execution (git/gh) including push and PR creation.\n
- Sanitization: No sanitization or escaping of ingested data is described.
Audit Metadata