skills/github/gh-aw/gh-agent-session/Gen Agent Trust Hub

gh-agent-session

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches and installs the agent-task extension from the official GitHub repository (github/agent-task), which is a verified vendor source.- [COMMAND_EXECUTION]: Uses the gh CLI to automate coding sessions based on natural language instructions, which is the documented and intended behavior of the extension.- [CREDENTIALS_UNSAFE]: Correctly identifies Personal Access Tokens (PATs) as required for operation and recommends storing them in repository secrets (COPILOT_GITHUB_TOKEN) to prevent credential leakage.- [PROMPT_INJECTION]: The skill ingests natural language task descriptions to generate code changes. This inherent indirect prompt injection surface is documented with appropriate mitigation strategies, including manual review of generated pull requests and the use of the GitHub Agentic Workflows safe-outputs framework.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 01:16 PM