See your OpenClaw agents think, in real time.
Real-time observability for sessions, channels, usage, memory, logs, and flow behavior.
- Zero-config startup on local machines
- Unified operational visibility across your OpenClaw runtime
- Lightweight architecture (single-file dashboard, no frontend build chain)
- Works well for local development and LAN dashboards
curl -fsSL https://raw.githubusercontent.com/plgonzalezrx8/clawtelemetry/main/install.sh | bashirm https://raw.githubusercontent.com/plgonzalezrx8/clawtelemetry/main/install.ps1 | iexcurl -fsSL https://raw.githubusercontent.com/plgonzalezrx8/clawtelemetry/main/install.cmd -o install.cmd && install.cmd && del install.cmdclawtelemetry --host 0.0.0.0 --port 8900Open:
http://localhost:8900http://<your-lan-ip>:8900(if LAN access is enabled and your firewall allows it)
Common options:
clawtelemetry --port 9000
clawtelemetry --host 127.0.0.1
clawtelemetry --workspace ~/mybot
clawtelemetry --name "Alice"clawtelemetry update- Pulls the latest GitHub Release.
- Automatically restarts a running background ClawTelemetry service after successful update.
- For CI/offline environments, disable periodic checks with
CLAWTELEMETRY_DISABLE_UPDATE_CHECK=1.
git clone https://github.com/plgonzalezrx8/clawtelemetry.git
cd clawtelemetry
python3 dashboard.py --host 0.0.0.0 --port 8900- Overview
- Usage
- Sessions
- Crons
- Logs
- Memory
- Transcripts
- Flow
| Overview | Usage | Memory |
|---|---|---|
![]() |
![]() |
![]() |
| Brain | Overview (v3) | Usage (v3) |
|---|---|---|
![]() |
![]() |
![]() |
- Architecture: ARCHITECTURE.md
- Cloud and remote-access notes: docs/CLOUD_TESTING.md
- Contributing: CONTRIBUTING.md
- Security policy: SECURITY.md
- Internal engineering docs: DEV-DOCS/README.md
MIT







