RepoRadar
Leave a Tip
Item detail

latitude-dev/latitude-llm

Latitude is an MIT-licensed open-source issue-detection and observability platform for AI agents and LLM apps — 'Sentry but for agents.' 4,146 stars, ships issue-centric tracing (failed traces grouped into tracked issues with status, size, and trend), human-aligned evals built from team judgments, agent-native traces with full multi-turn + tool-call context, and semantic search over 100% of traces

Open source →Back to radar
Score7.9
Popularity78.0
Risknone
TierSilver
Score breakdown
Usefulness8.0
Novelty8.0
Momentum8.0
Maturity7.4
Open-source/build8.4
Evidence7.2
Workflow potential9.0
Setup ease8.8

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for engineering teams running production AI agents that need to answer 'what will break next' before users notice. Install the @latitude-data/telemetry SDK, instrument your agent, and let Latitude group the failures into tracked issues with alignment scores that track drift from human judgment.

Who should use it

engineering teams running production AI agents that need to detect failures before users noticeplatform teams standardizing on a Sentry-like workflow for AI agent observability instead of bolting it onto Datadog or New Relicteams that want evals built from human judgment with alignment scores that track drift over timeagent developers debugging multi-turn tool-call failures that don't surface in single-turn eval frameworksorganizations that need agent observability with a self-hostable open-source core

Who should skip it

Skip if the source link, docs, or setup requirements do not match your workflow.

Risk explanation

No inherent user-impacting risk is flagged from the captured evidence.

Evidence links

Closest alternatives / related signals

evaluationobservabilityagent-tracingissue-trackingsentry-for-agentshuman-aligned-evalssemantic-searchself-hostable