Mandoline
This Mandoline MCP server, developed by Mandoline AI, provides AI assistants with direct access to Mandoline's LLM evaluation platform through a comprehensive set of tools for creating and managing custom metrics, running evaluations on prompt-response pairs, and retrieving evaluation results. Built with TypeScript and Express, it features session management with automatic cleanup, API key-based authentication, and tools for both individual and batch operations on metrics and evaluations. The implementation includes automatic environment context injection (client info, model names) and content hashing for evaluation tracking, making it particularly useful for AI development teams who need to systematically evaluate and improve their language models through conversational interfaces rather than traditional dashboards.
Composite of vulnerability cleanliness, spec conformance, provenance, stability, and usage signals — scanned and weighted by Cognium. Human and agent signals are tracked separately. Last scanned 2026-09-02.
Scan details: Circle-IR · 2026-09-02 · Appeal
View full trust & usage report →Metadata
- Version
- 1.0.0
- Skill type
- atomic
- Execution layer
- mcp-remote
- Category
- ai-ml
- Source
- PulseMCP
- Repository
- github.com/mandoline-ai/mandoline-mcp-server
- Author type
- human
- Last scanned
- 2026-09-02
- Updated
- 2026-09-02
Use via MCP
Resolve Mandoline from your agent
Streamable HTTP transport at https://api.skillsregistry.net/mcp. No auth for read tools. Discovery: .well-known/mcp.json.
One command in your shell — Claude Code wires it up and verifies the connection. Run /mcp in any session to confirm.
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp--scope user for --scope project to commit it to .mcp.json.