github-issues

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes data from GitHub issues and comments, which are untrusted external sources.\n
  • Ingestion points: Content is fetched using MCP tools like mcp__github__issue_read and search functions, or via gh api GET requests.\n
  • Boundary markers: No specific delimiters or security instructions are provided to the agent to mitigate the risk of embedded instructions in the fetched data.\n
  • Capability inventory: The skill allows the agent to execute write actions on GitHub repositories using mcp__github__issue_write, mcp__github__add_issue_comment, and the gh CLI.\n
  • Sanitization: The instructions do not include specific measures for sanitizing or validating ingested data before use.\n- [COMMAND_EXECUTION]: The skill instructions frequently utilize the gh CLI for administrative tasks and API interactions.\n
  • Evidence: SKILL.md and several reference files contain numerous examples of shell commands involving gh api, gh label, and other CLI utilities to manage GitHub resources. This is the core functionality of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 11:36 AM
Security Audit — agent-trust-hub — github-issues