avito-mcp

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from the external Avito API, which could potentially contain malicious instructions designed to influence the agent's behavior.
  • Ingestion points: Data enters the context via the avito_call_method tool which retrieves information from the Avito marketplace API.
  • Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded in the API responses.
  • Capability inventory: The skill has the capability to perform network requests (API calls) and write to the local file system (~/.marketplace-mcp/cabinets.json).
  • Sanitization: There is no mention of sanitization or filtering of the content received from the API before it is processed by the agent.
  • [EXTERNAL_DOWNLOADS]: The installation process involves downloading and executing code from the author's public GitHub repository using uvx.
  • Evidence: Fetches the server package from https://github.com/ilyautov/avito-mcp-ru during installation and configuration.
  • [CREDENTIALS_UNSAFE]: The skill requires sensitive API credentials (AVITO_CLIENT_ID and AVITO_CLIENT_SECRET) to function.
  • Evidence: Instructions guide the user to provide these secrets via environment variables or by storing them in a local JSON file (~/.marketplace-mcp/cabinets.json) with restricted permissions (600).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 02:16 PM
Security Audit — agent-trust-hub — avito-mcp