Traces, scores, and evaluates production AI agents via OpenTelemetry and SDKs, groups failures into signals, and dispatches coding agents over MCP to fix them.
Use it now
Keep it with up to 9 other tools you're setting up, then share them all as one link.
Try this first
Ask your assistant, with Latitude connected:
- Trace Ingestion
Use Latitude to capture multi-turn sessions, tool calls, and execution paths: ‹what you need›
- Agent Score
Use Latitude to compute a daily 0 to 100 project score: ‹what you need›
- Automated Evaluations
Use Latitude to score sessions: ‹what you need›
Connect it over MCP
Connect Latitude as an MCP server, following its setup instructions at https://docs.latitude.so
Its page: docs.latitude.so
Install the CLI
Install it as its own instructions say.
Its page: docs.latitude.so
Call the API
Auth An API key from Latitude
Its page: docs.latitude.so
Compare
Press Compare to add it to the tray at the bottom of the window, then pick up to 2 others anywhere on the site.
- Access
- APICLIMCP
- Auth
- API Key
- Skills
- 4
Skills
Trace Ingestion
Captures multi-turn sessions, tool calls, and execution paths through TypeScript and Python SDKs or any OpenTelemetry exporter.
Agent Score
Computes a daily 0 to 100 project score across outcome, reliability, cost, speed, and safety from real production sessions.
Automated Evaluations
Scores sessions with condition sets, LLM judges, or custom JavaScript scripts, and replays regression datasets against fixed failures.
Agent Dispatch
Wakes a coding agent with sample traces when a signal needs follow-up, via MCP, Linear, or webhooks, so it can investigate and open a PR.
