bugfix
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches external, untrusted data from GitHub issues using the
gh issue viewcommand. This content is then passed to other AI agents (thetest-engineer) to drive reproduction and verification strategies. A malicious actor could embed instructions within a GitHub issue body to manipulate the agent's behavior during the bug-fixing process. - [COMMAND_EXECUTION]: The skill provides instructions for executing shell commands, including directory creation (
mkdir) and GitHub CLI operations (gh). These are used within the intended scope of repository management and issue tracking.
Audit Metadata