github-triage

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data from GitHub issues and comments during the triage workflow.\n
  • Ingestion points: Issue bodies and comments are retrieved using the gh tool in the context gathering phase (SKILL.md, Step 1).\n
  • Boundary markers: The instructions do not specify any boundary markers or delimiters to isolate untrusted content from the agent's core instructions.\n
  • Capability inventory: The agent has access to gh, git, and general shell command execution for bug reproduction and file modification.\n
  • Sanitization: No sanitization or validation of the issue content is performed before it is processed by the agent.\n- [COMMAND_EXECUTION]: The 'Bug reproduction' section in SKILL.md (Step 3) explicitly directs the agent to 'execute commands' based on the steps provided by an issue reporter, providing a direct path for untrusted external input to trigger command execution on the host environment.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 02:59 PM
Security Audit — agent-trust-hub — github-triage