github-workflow-automation

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches and executes automation logic from vendor-owned packages ruv-swarm and claude-flow using the npx utility.
  • [COMMAND_EXECUTION]: Utilizes the GitHub CLI (gh) and Git to perform repository operations such as creating issues, managing pull requests, and viewing workflow logs.
  • [PROMPT_INJECTION]: Identifies a surface for indirect prompt injection due to the processing of external repository data.
  • Ingestion points: The skill reads external content through gh pr view for pull request files and metadata (line 242) and monitors workflow run outputs (line 197).
  • Boundary markers: No specific delimiters or instructions to ignore embedded directives within processed content were observed in the workflow templates.
  • Capability inventory: Capabilities include the ability to create and comment on GitHub issues and pull requests, as well as orchestrating multi-agent tasks via the ruv-swarm framework.
  • Sanitization: Content is parsed using jq to extract specific fields, but the extracted text is subsequently used in automated coordination and reporting tasks without additional filtering or escaping.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 05:44 AM