NYC

github-repo-architect

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (LOW): The skill utilizes powerful GitHub CLI (gh) commands to create repositories, clone code, and modify organization-wide workflows. While appropriate for its stated purpose, these capabilities should be monitored as they allow the agent to make significant changes to codebases.
  • [EXTERNAL_DOWNLOADS] (LOW): The skill templates include dependencies on ruv-swarm, an external package not listed among trusted sources. Users should verify the integrity of this package before deployment.
  • [PROMPT_INJECTION] (LOW): The skill exhibits an indirect prompt injection surface by reading untrusted data from package.json files and repository search results without explicit boundary markers.
  • Ingestion points: cat package.json, gh search repos, gh repo list.
  • Boundary markers: Absent; the skill does not use specific delimiters or instructions to ignore embedded commands in the files it reads.
  • Capability inventory: Extensive, including git push, gh repo create, and gh pr create which could be misused if the agent is influenced by malicious data.
  • Sanitization: Absent; the skill processes file contents and repository metadata directly into its workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:27 PM