Hugging Face MCP
Discover models and datasets and access documentation on the Hugging Face Hub from an agent.
🛡️ AgentReady threat assessment
MAESTRO 7-layer threat model + OWASP AIVSS risk score for Hugging Face MCP, derived from its capabilities.
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology. Are you the vendor? Factual corrections are free.
Overview
The Hugging Face connector lets agents search and explore models, datasets, and Spaces on the Hub and read model documentation and capabilities. It exposes read access to public Hub metadata and, with a token, private assets. Retrieved model cards and dataset READMEs are untrusted text, and any token grants access to a user's private repos.
Key features
- Model and dataset search
- Model card and docs retrieval
- Spaces discovery
- Optional authenticated access
Use cases
- Find and compare ML models
- Explore datasets for a task