Features
Full agent tracing: every tool call, LLM hop, decision branch
One-line instrumentation for major agent frameworks
Works with any LLM provider out of the box
SOTA uncertainty detection over long trajectories
LLM-as-judge scoring with structured feedback
Session-level evaluation (not just isolated traces)
Automated monitoring with scheduled eval runs
Alerting on metric regressions across agent versions
PandaProbe Skill for coding agents (Claude Code, Cursor, Codex)
PandaProbe CLI for terminal-based trace/eval management
Open-source self-hosting (Apache 2.0)
Data retention management (Startup+ tiers)
Custom SSO (Enterprise tier)
Pay-as-you-go overages beyond included quotas
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
