skill-architecture
Pass
Audited by Gen Agent Trust Hub on Apr 26, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill synchronizes documentation and templates from trusted sources, including official Anthropic repositories and verified developer resources.
- [COMMAND_EXECUTION]: Provides structured templates for executing local development tools such as git, uv, bun, and mise for project management and validation.
- [PROMPT_INJECTION]: While the skill ingests user requirements to generate new instructions (Indirect Prompt Injection surface), it mitigates risk through mandatory architectural templates, validation scripts, and human-in-the-loop checkpoints.
- [REMOTE_CODE_EXECUTION]: Executes local, bundled scripts for initializing and packaging skills; no untrusted remote script execution or dangerous piping patterns were detected.
Audit Metadata