github-workflow-automation

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill extensively uses the GitHub CLI (gh) and Git to manage repository operations, including creating issues, commenting on pull requests, and viewing run logs.
  • [EXTERNAL_DOWNLOADS]: The skill relies on npx to fetch and execute ruv-swarm and claude-flow@alpha. These are vendor-owned resources associated with the author 'ruvnet' and are used for core functionality.
  • [REMOTE_CODE_EXECUTION]: The use of npx to run remote packages like ruv-swarm constitutes dynamic execution of external code. While these are vendor resources, the mechanism involves fetching and running code at runtime.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection due to its interaction with untrusted external data.
  • Ingestion points: The skill ingests data from pull request views (gh pr view), workflow run details (gh run view), and repository codebases.
  • Boundary markers: There are no explicit delimiters or 'ignore' instructions implemented in the provided command examples to prevent the agent from obeying instructions embedded in PR descriptions or issue bodies.
  • Capability inventory: The skill has broad capabilities, including executing shell commands via npx, creating GitHub issues, and posting pull request comments.
  • Sanitization: The provided examples do not demonstrate any sanitization, filtering, or validation of the data retrieved from GitHub before it is processed by the AI swarm.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 07:39 PM