issues-triage
Pass
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from GitHub issues and comments which may contain malicious instructions designed to exploit the agent's capabilities.
- Ingestion points: Issue bodies and comments are retrieved from GitHub using the
gh issue viewcommand. - Boundary markers: The skill includes specific instructions to treat external text as data and ignore embedded commands, scripts, or links.
- Capability inventory: The agent can execute shell commands (
git,gh) and perform file system writes to create local task files. - Sanitization: Instructions require the agent to ignore potentially dangerous content like links or attachments within issues and to flag security vulnerabilities to the user.
- [COMMAND_EXECUTION]: The skill executes local commands to interact with the repository and the GitHub API.
- Commands used:
git switch,git pull,gh issue edit, andgh issue commentare used to manage project state and GitHub issues. - User oversight: A human-in-the-loop step is mandatory, requiring the user to approve or modify all proposed actions before they are executed.
Audit Metadata