triage
Warn
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: MEDIUMINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and process the full body and comments of issues and pull requests provided by external reporters. * Ingestion points: Context gathering in SKILL.md (Step 1) reads all issue/PR content. * Boundary markers: No explicit delimiters or instructions to ignore malicious payload within the ingested text are present. * Capability inventory: The agent is instructed to execute commands (Step 3), write files (Step 5), and post comments to trackers. * Sanitization: No sanitization of input text is mentioned.
- [COMMAND_EXECUTION]: The skill directs the agent to 'run the relevant tests or commands' from external pull requests to verify their claims. * Evidence: SKILL.md (Step 3: Verify the claim) instructs the agent to check out and run code provided by external contributors. * Risk: Malicious pull requests could include harmful commands in their test suites or verification scripts.
- [REMOTE_CODE_EXECUTION]: Verifying pull requests involves fetching and running code from external, untrusted branches. * Evidence: SKILL.md (Step 3) instructions for the pull request verification workflow which entails checking out and executing external code.
Audit Metadata