Atla MCP Server provides a standardized interface for LLMs to interact with the Atla API for state-of-the-art LLMJ evaluation. It offers tools for evaluating LLM responses against single or multiple criteria, returning both numerical scores and textual critiques. Built with Python 3.11 and distributed under the MIT license, this server enables AI assistants to leverage Atla's evaluation models for assessing response quality across various dimensions. The implementation supports multiple connection methods including OpenAI Agents SDK, Claude Desktop, and Cursor, making it accessible for developers working with different MCP clients.
Cognium trust score
50%
Tier
Unverified
Composite of vulnerability cleanliness, spec conformance, provenance, stability, and usage signals — scanned and weighted by Cognium. Human and agent signals are tracked separately.
Returns 7 tools: search_skills, get_skill, list_leaderboard, get_trust_breakdown, resolve_composition, plus the ChatGPT-connector search and fetch. Every tool is annotated read-only.
Resolve this skill directly via MCP tools/call get_skill.