Features
Live Monitor real-time agent activity over WebSocket
Cost & Token Analytics per model, agent, and time range
Sessions table with cost, tokens, cache hit rate, context pressure, errors
Session Timeline Gantt chart of LLM calls and tool invocations
Agents overview showing health, cost, error rate, context pressure
Security Audit scanning for credential leaks, prompt injection, shell commands
Context Breakdown visualization of context window usage
Cron Jobs auto-detection with run history logs
Tool Profiler latency distribution and failure rates
Deep Turns detection for excessive tool-call sequences
Cache Trace pipeline flow visualization and diff between LLM calls
Memory viewer for agent state inspection
Cache hit rate analytics for token savings
i18n support for English and Simplified Chinese
CLI-based zero-config setup with `npx claw-lens-cli`
Trace visibility for agent steps (prompts, retrievals, tool calls, outputs)
LLM-as-judge evaluation for relevance, toxicity, quality scoring
Dataset creation from traces for reproducible testing
Experiment management and regression benchmarking
Built-in Prompt IDE for iterative prompt optimization
Self-hosted deployment on local, Docker, Kubernetes
Phoenix Cloud managed hosting option
Vendor-agnostic support for any model/framework
Native OpenTelemetry integration
OpenInference specification for LLM telemetry
Human annotation and automated labeling
Ghost trajectories to simulate alternative agent paths
Eval-as-you-test for early quality feedback
One-click integration with LlamaIndex
