AI Observer
Open Source & Self-Hosted

Unified Local Observability for AI Coding Assistants

Track costs, debug traces, and monitor all your AI tools in one privacy-first dashboard. Your data stays on your machine.

Works with your favorite AI tools

Everything You Need

Powerful features to understand and optimize your AI tool usage

Multi-Tool Support

Works with Claude Code, Gemini CLI, and OpenAI Codex CLI. Monitor all your AI coding assistants from one unified dashboard.

Real-Time Dashboard

Live WebSocket updates as telemetry arrives. Customizable drag-and-drop widgets with multiple visualization types.

Cost Tracking

Embedded pricing data for 67+ models across Claude, Codex, and Gemini. Understand exactly what you're spending.

Historical Import

Import past sessions from local JSONL/JSON files with automatic cost calculation. Analyze your complete AI tool history.

DuckDB Analytics

Powered by DuckDB for instant queries on large datasets. Fast aggregations, filtering, and time-series analysis.

Privacy First

All data stays on your machine. Self-hosted with zero external dependencies—complete control over your telemetry.

Single Binary

One ~54MB executable with embedded frontend. No external dependencies, no configuration—just run and go.

OTLP Native

Standard OpenTelemetry Protocol ingestion via HTTP/JSON and HTTP/Protobuf. Works with any OTLP-compatible tool.

Data Export

Export telemetry to portable Parquet files with optional DuckDB views database. Archive, share, or analyze externally.

How It Works

Get up and running in minutes with three simple steps

Architecture
AI Coding Tools Claude Code Gemini CLI OpenAI Codex Other OTLP Tools OTLP/HTTP (Protobuf/JSON) AI Observer Server OTLP Receiver (HTTP/Protobuf) Cost Engine (67+ AI models) WebSocket (Real-time) REST API (Query Data) DuckDB (Analytics) Real-Time Dashboard Metrics Widgets Logs Viewer Traces Explorer
01

Start AI Observer

Run AI Observer locally via Docker, Homebrew, or binary. It starts a local server that listens for telemetry data.

02

Configure Your AI Tools

Point your AI coding assistants to send OpenTelemetry data to AI Observer. Simple environment variable configuration.

03

Monitor & Analyze

Open the dashboard to see real-time metrics, traces, and logs. Track costs, debug issues, and optimize usage.

Quick Start

Get AI Observer running in under a minute

Docker Installation
# Pull and run AI Observer
docker run -d \
  --name ai-observer \
  -p 4318:4318 \
  -p 8080:8080 \
  -v ai-observer-data:/data \
  tobilg/ai-observer:latest

# Dashboard: http://localhost:8080
# OTLP endpoint: http://localhost:4318

Configure Your AI Tools

Point your AI coding assistant to send telemetry to AI Observer:

Claude Code Setup
# Add to ~/.bashrc or ~/.zshrc
export CLAUDE_CODE_ENABLE_TELEMETRY=1
export OTEL_METRICS_EXPORTER=otlp
export OTEL_LOGS_EXPORTER=otlp
export OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
export OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4318
export OTEL_METRIC_EXPORT_INTERVAL=10000
export OTEL_LOGS_EXPORT_INTERVAL=5000
export OTEL_LOG_USER_PROMPTS=1

See It in Action

A beautiful, powerful dashboard for all your AI observability needs

Dashboard Overview

Dashboard Overview

Real-time metrics and KPIs at a glance

Cost Analytics

Cost Analytics

Track spending across all AI models

Trace Explorer

Trace Explorer

Debug and analyze AI tool interactions

Logs Viewer

Logs Viewer

Search and filter telemetry logs

Free & Open Source

No subscriptions, no limits, no strings attached

MIT License
$0 /forever

Self-hosted on your infrastructure

  • Unlimited AI tools
  • Unlimited telemetry data
  • Real-time dashboard
  • Cost tracking for 67+ models
  • Trace & log analysis
  • Self-hosted, privacy-first
  • No usage limits
  • Community support

Join the community

Star on GitHub

Take Control of Your AI Tool Usage

Stop flying blind. Start understanding exactly how your AI coding assistants work, what they cost, and how to optimize them.