git-setup-repo

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses standard repository inspection and setup tools (Git, Ripgrep, Node.js package managers) to analyze and configure the development environment.\n- [SAFE]: The skill provides templates for local development services (Docker Compose) that include default credentials (e.g., 'root'/'pass'). These are intended for isolated local development and do not represent a production security risk.\n- [COMMAND_EXECUTION]: The skill executes routine administrative commands for repository initialization, dependency management, and local service orchestration via Docker and shell scripts.\n- [PROMPT_INJECTION]: The skill processes untrusted repository files (e.g., AGENTS.md, README.md) to build a setup brief, which represents an indirect prompt injection surface.\n
  • Ingestion points: Audit phase reads AGENTS.md, README.md, package.json, and .github workflows from the target repository (as described in SKILL.md).\n
  • Boundary markers: Absent. The skill encourages reading arbitrary local documentation to infer requirements.\n
  • Capability inventory: File system writing (cp), package execution (npm install), and shell script execution (provision.sh) across multiple files.\n
  • Sanitization: Absent. Data is used directly to propose repository modifications.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 12:31 AM