anthropic-api-tools-mcp
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a comprehensive documentation repository and contains no executable code or malicious instructions. It provides reference material for the Anthropic Claude API, including JSON Schema definitions, API endpoint signatures, and deployment guides.
- [EXTERNAL_DOWNLOADS]: The documentation identifies trusted external sources for tools and configuration, such as the official Anthropic GitHub repository (
github.com/anthropics/skills) and the Anthropic public container registry (us-docker.pkg.dev/anthropic-public-registry). These references are documented neutrally for installation and deployment purposes. - [INDIRECT_PROMPT_INJECTION]: The skill describes attack surfaces associated with processing external data (e.g., via web search or text editing tools) and provides explicit security guidance on mitigating these risks, such as treating tool outputs as untrusted and using strict allowlists for command execution.
- [COMMAND_EXECUTION]: While the documentation includes examples of bash and python commands, these are clearly marked as illustrative samples for users to implement in their own isolated environments and are not executed by the skill itself.
Audit Metadata