anthropic-connector-submit
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely informational and contains no executable code, scripts, or automated background tasks. All operations described are intended for the user to perform manually via external forms.
- [INDIRECT_PROMPT_INJECTION]: While the skill instructions facilitate the processing of user-provided server metadata (names, descriptions, tags), there is no evidence of an attack surface for code execution or unauthorized data access. The skill lacks tools for writing to the filesystem or making arbitrary network requests.
- [EXTERNAL_DOWNLOADS]: The skill links to a Google Form and the Smithery website. These are well-known services relevant to the skill's stated purpose of MCP server distribution and do not involve the downloading of untrusted scripts or binary payloads.
- [CREDENTIALS_UNSAFE]: The skill appropriately advises users to share test credentials via secure links (e.g., 1Password) rather than plain text, demonstrating adherence to security best practices for the submission process.
Audit Metadata