GitHub - latitude-dev/latitude-llm: Latitude is the open-source ai monitoring platform.

Issue detection for AI Agents

Sentry, but for agents and LLMs.

Website  ·   Docs  ·   Slack  ·   X

🌈 Why Latitude?

Latitude shows you what will break next in your AI Agent and helps you fix it before users notice.

• Issue-centric: failed traces grouped into tracked issues, with status, size, and trend.
• Human-aligned evals: evals built automatically from your team's judgments, with an alignment score that tracks drift from human judgment over time.
• Agent-native traces: multi-turn sessions, tool calls, and full execution paths in one view.
• Semantic search: find any trace by meaning, exact matches, or roughly similar sentences. No sampling, 100% of traces are searchable.

📚 Table of contents

• Quick start
• Integrations
• With Claude Code
• Roadmap
• Community
• Contributing
• License
• Links

⚡ Quick start

Sign up at latitude.so and grab your API key and project slug.

Recommended: ask your coding agent

Paste this prompt into Claude Code, Cursor, Windsurf, Codex, OpenCode, or another coding agent:

Read the Latitude Telemetry AI skill from https://raw.githubusercontent.com/latitude-dev/skills/refs/heads/main/skills/latitude-telemetry/SKILL.md and add tracing to this application.

Manual TypeScript setup

npm install @latitude-data/telemetry

This example uses OpenAI; replace it with the LLM SDK your app already imports.

import { Latitude } from "@latitude-data/telemetry";
import OpenAI from "openai";

const latitude = new Latitude({
  apiKey: process.env.LATITUDE_API_KEY!,
  project: process.env.LATITUDE_PROJECT_SLUG!,
  instrumentations: { openai: OpenAI },
});

const client = new OpenAI();

await client.chat.completions.create({
  model: "gpt-4o",
  messages: [{ role: "user", content: "Hello" }],
});

await latitude.shutdown();

Every supported LLM call now shows up as a trace in Latitude. Use capture() at request, conversation, or agent boundaries when you want to add user IDs, session IDs, tags, or metadata.

Python and any OpenTelemetry-compatible runtime are also supported. Full setup, provider guides, and OTel passthrough are in the Start tracing guide.

🔌 Integrations

Latitude is provider-agnostic. Telemetry works out of the box with most model providers and frameworks (OpenAI, Anthropic, Bedrock, Vercel AI SDK, LangChain, and more), plus any OTLP-compatible backend.

See the full integration list for setup instructions.

✳️ With Claude Code

Building inside Claude Code? We have a dedicated package that captures full session transcripts as traces. Check out docs.

npx -y @latitude-data/claude-code-telemetry install

Works in the terminal, the Desktop app, and IDE extensions.

Roadmap

• OTel ingest, traces, sessions
• Issue discovery + clustering
• Issue-to-eval generation with alignment optimization
• Eval alignment tracking (MCC, coverage)
• Semantic search: find traces by meaning
• Potential issues: track failure modes you already know about
• Stable v2 release

👥 Community

Join the Slack community to ask questions, share feedback, and show what you're building.

🤝 Contributing

Contributions are welcome. If you want to help, join the Slack community, open an issue, or submit a pull request.

📄 License

Latitude is licensed under the MIT License.

🔗 Links

• Home page
• Documentation
• Slack community
• X / Twitter

Made with love by the Latitude Team