> ## Documentation Index > Fetch the complete documentation index at: https://overmind-b84ae13c-codex-workshop-queue-capacity.mintlify.site/llms.txt > Use this file to discover all available pages before exploring further. # Introduction > Overmind improves production AI agents with their own data: traces in; datasets, evals, optimisation runs and trained models out. export const Ascii = ({lines, cols}) =>
line.length)) }}> 
{lines.join("\n")}
; Overmind ingests OpenTelemetry traces, scores them as they arrive and turns them into datasets. Those datasets drive evals, optimisation runs and training runs, and every result is benchmarked against the agent you ship today. Use the [hosted Console](https://console.overmindlab.ai) or [self-host](/platform/self-hosting) the same code with Docker Compose. The platform is AGPL-3.0 and the SDK is MIT; see [Licensing](/platform/licensing). Everything speaks standard protocols: OTLP/HTTP for ingest, REST JSON for the API, [MCP](/platform/mcp) for coding agents and OpenAI-compatible serving for models. The Python SDK instruments OpenAI, Anthropic, Gemini, Agno and LangChain, and any OpenTelemetry exporter can send to the same endpoint. What Overmind does: * [Capability scan](/core/capabilities): your coding agent maps the repository into capabilities * [Live scoring](/agent-testing/trace-scoring): every trace is carved into task executions and judged against the capability's eval set * [Datasets](/core/datasets): traces and files become eval and training data, shaped by the dataset's own agent * [Eval](/agent-testing/eval): variants measured against a baseline with typed scores * [Optimiser](/agent-testing/optimisers): candidate code changes replayed in your repository and scored; the winning diff is yours to apply * [Training](/models/training): fine-tuned models you own, trained on Modal from your data * [Inference](/models/inference): an OpenAI-compatible API for your trained models and frontier models * [MCP](/platform/mcp) and [CLI](/platform/cli): the same workflows from Cursor, Claude Code, OpenCode or Codex ## Guides One pasted prompt: SDK, MCP, capability scan, first trace. Docker Compose, the environment reference, backends and billing modes. OTLP ingest, span model, attribute mapping, the executions view. `init()`, the run boundary, decorators, tasks, runtime expectations. Local discovery, `overmind sync`, tasks, telemetry attribution. Source, cells, versions, contracts, the workshop agent. Evaluator kinds, claims, task bindings, eval runs. Units, tasks, verdicts, execution and session scores. The client-driven loop, harness and backtesting modes, the winning diff. Setup per coding agent, the tool catalog, prompts, resources. ## Models & API Runs, the model catalog, loss curves, the baseline benchmark. Serving lifecycle, the capability alias, `/api/v1/chat/completions`. Auth, endpoint map, conventions, Swagger. `overmind init`, `sync`, `dataset`, `optimise`, `model`. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.