claude-code-agent-architecture
Warn
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides installation steps that require cloning a third-party GitHub repository (https://github.com/sanbuphy/learn-coding-agent.git) and running npm install or bun install. Executing these commands results in the download and potential execution of arbitrary code from an unverified source.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user queries to provide information about AI agent architecture, creating a surface for indirect prompt injection.
- Ingestion points: User triggers regarding Claude Code architecture and internal mechanisms defined in SKILL.md.
- Boundary markers: None present.
- Capability inventory: The skill includes instructions for git clone and package installation (npm install, bun install).
- Sanitization: No input validation or content sanitization is documented for the processing of external repository data.
Audit Metadata