# PandaProbe

## Docs

- [Create Batch Eval Run](https://docs.pandaprobe.com/api-reference/evaluations/create-batch-eval-run.md): Create an eval run for an explicit list of trace IDs.
- [Create Batch Session Eval Run](https://docs.pandaprobe.com/api-reference/evaluations/create-batch-session-eval-run.md): Create a session eval run for explicit session IDs.
- [Create Eval Run](https://docs.pandaprobe.com/api-reference/evaluations/create-eval-run.md): Create a filtered eval run.
- [Create Monitor](https://docs.pandaprobe.com/api-reference/evaluations/create-monitor.md): Create an evaluation monitor that spawns eval runs on a recurring schedule.
- [Create Session Eval Run](https://docs.pandaprobe.com/api-reference/evaluations/create-session-eval-run.md): Create a filter-based session eval run.
- [Create Trace Score](https://docs.pandaprobe.com/api-reference/evaluations/create-trace-score.md): Manually create a trace score.
- [Delete Eval Run](https://docs.pandaprobe.com/api-reference/evaluations/delete-eval-run.md): Delete an eval run.
- [Delete Monitor](https://docs.pandaprobe.com/api-reference/evaluations/delete-monitor.md): Delete an evaluation monitor. Spawned runs are preserved.
- [Delete Session Eval Run](https://docs.pandaprobe.com/api-reference/evaluations/delete-session-eval-run.md): Delete a session eval run.
- [Delete Session Score](https://docs.pandaprobe.com/api-reference/evaluations/delete-session-score.md): Delete a single session score.
- [Delete Trace Score](https://docs.pandaprobe.com/api-reference/evaluations/delete-trace-score.md): Delete a single trace score.
- [Get Available Metrics](https://docs.pandaprobe.com/api-reference/evaluations/get-available-metrics.md): List all registered evaluation metrics.
- [Get Available Providers](https://docs.pandaprobe.com/api-reference/evaluations/get-available-providers.md): List LLM providers and their availability.
- [Get Available Session Metrics](https://docs.pandaprobe.com/api-reference/evaluations/get-available-session-metrics.md): List all registered session evaluation metrics.
- [Get Eval Run](https://docs.pandaprobe.com/api-reference/evaluations/get-eval-run.md): Get full eval run detail.
- [Get Eval Run Template](https://docs.pandaprobe.com/api-reference/evaluations/get-eval-run-template.md): Return a pre-filled eval run template for a single metric.
- [Get Monitor](https://docs.pandaprobe.com/api-reference/evaluations/get-monitor.md): Get evaluation monitor detail.
- [Get Scores For Run](https://docs.pandaprobe.com/api-reference/evaluations/get-scores-for-run.md): List all trace scores produced by a specific eval run.
- [Get Scores For Session](https://docs.pandaprobe.com/api-reference/evaluations/get-scores-for-session.md): Get all scores for a specific session.
- [Get Scores For Trace](https://docs.pandaprobe.com/api-reference/evaluations/get-scores-for-trace.md): Get the latest score per metric for a specific trace.
- [Get Session Eval Run](https://docs.pandaprobe.com/api-reference/evaluations/get-session-eval-run.md): Get full session eval run detail.
- [Get Session Score Analytics Distribution](https://docs.pandaprobe.com/api-reference/evaluations/get-session-score-analytics-distribution.md): Histogram of session score values for a metric.
- [Get Session Score Analytics Summary](https://docs.pandaprobe.com/api-reference/evaluations/get-session-score-analytics-summary.md): Aggregated session score summary per metric.
- [Get Session Score Analytics Trend](https://docs.pandaprobe.com/api-reference/evaluations/get-session-score-analytics-trend.md): Time series of average session scores by metric.
- [Get Session Score Comparison](https://docs.pandaprobe.com/api-reference/evaluations/get-session-score-comparison.md): Leaderboard: latest score per session for a metric, sorted by value.
- [Get Session Score History](https://docs.pandaprobe.com/api-reference/evaluations/get-session-score-history.md): Score evolution for a specific session across re-evaluations over time.
- [Get Session Scores For Run](https://docs.pandaprobe.com/api-reference/evaluations/get-session-scores-for-run.md): List all session scores produced by a specific eval run.
- [Get Trace Score Analytics Distribution](https://docs.pandaprobe.com/api-reference/evaluations/get-trace-score-analytics-distribution.md): Histogram of trace score values for a metric.
- [Get Trace Score Analytics Summary](https://docs.pandaprobe.com/api-reference/evaluations/get-trace-score-analytics-summary.md): Aggregated trace score summary per metric.
- [Get Trace Score Analytics Trend](https://docs.pandaprobe.com/api-reference/evaluations/get-trace-score-analytics-trend.md): Time series of average trace scores by metric.
- [List Eval Runs](https://docs.pandaprobe.com/api-reference/evaluations/list-eval-runs.md): List trace eval runs (summary view).
- [List Monitor Runs](https://docs.pandaprobe.com/api-reference/evaluations/list-monitor-runs.md): List eval runs spawned by a specific monitor.
- [List Monitors](https://docs.pandaprobe.com/api-reference/evaluations/list-monitors.md): List evaluation monitors.
- [List Session Eval Runs](https://docs.pandaprobe.com/api-reference/evaluations/list-session-eval-runs.md): List session eval runs.
- [List Session Scores](https://docs.pandaprobe.com/api-reference/evaluations/list-session-scores.md): List session scores with filters.
- [List Trace Scores](https://docs.pandaprobe.com/api-reference/evaluations/list-trace-scores.md): List trace scores (summary view) with comprehensive filters.
- [Pause Monitor](https://docs.pandaprobe.com/api-reference/evaluations/pause-monitor.md): Pause an evaluation monitor. Idempotent if already paused.
- [Resume Monitor](https://docs.pandaprobe.com/api-reference/evaluations/resume-monitor.md): Resume a paused evaluation monitor. Idempotent if already active.
- [Retry Failed Eval Run](https://docs.pandaprobe.com/api-reference/evaluations/retry-failed-eval-run.md): Retry failed metrics from a completed eval run.
- [Retry Failed Session Eval Run](https://docs.pandaprobe.com/api-reference/evaluations/retry-failed-session-eval-run.md): Retry failed metrics from a completed session eval run.
- [Trigger Monitor](https://docs.pandaprobe.com/api-reference/evaluations/trigger-monitor.md): Force an immediate eval run from a monitor, ignoring cadence.
- [Update Monitor](https://docs.pandaprobe.com/api-reference/evaluations/update-monitor.md): Update an evaluation monitor.
- [Update Trace Score](https://docs.pandaprobe.com/api-reference/evaluations/update-trace-score.md): Manually edit a trace score.
- [Introduction](https://docs.pandaprobe.com/api-reference/introduction.md): PandaProbe REST API — authentication, base URL, conventions, and error handling.
- [Delete Session](https://docs.pandaprobe.com/api-reference/sessions/delete-session.md): Delete all traces (and cascaded spans) for a session.
- [Get Session](https://docs.pandaprobe.com/api-reference/sessions/get-session.md): Retrieve a single session with its full traces (including spans).
- [Get Session Analytics](https://docs.pandaprobe.com/api-reference/sessions/get-session-analytics.md): Time-series session statistics.
- [List Sessions](https://docs.pandaprobe.com/api-reference/sessions/list-sessions.md): List sessions for the current project.
- [Add Spans](https://docs.pandaprobe.com/api-reference/traces/add-spans.md): Add one or more spans to an existing trace (upsert).
- [Batch Delete](https://docs.pandaprobe.com/api-reference/traces/batch-delete.md): Delete multiple traces at once.
- [Batch Tags](https://docs.pandaprobe.com/api-reference/traces/batch-tags.md): Add or remove tags on multiple traces.
- [Delete Trace](https://docs.pandaprobe.com/api-reference/traces/delete-trace.md): Delete a trace and all its spans.
- [Get Analytics](https://docs.pandaprobe.com/api-reference/traces/get-analytics.md): Time-series analytics for traces.
- [Get Trace](https://docs.pandaprobe.com/api-reference/traces/get-trace.md): Retrieve a single trace with all its spans.
- [Ingest Trace](https://docs.pandaprobe.com/api-reference/traces/ingest-trace.md): Accept a trace payload for asynchronous persistence (upsert).
- [List Trace Users](https://docs.pandaprobe.com/api-reference/traces/list-trace-users.md): List unique user_ids with trace statistics.
- [List Traces](https://docs.pandaprobe.com/api-reference/traces/list-traces.md): List traces for the current project with filtering, sorting, and stats.
- [Update Span](https://docs.pandaprobe.com/api-reference/traces/update-span.md): Partially update a span on a trace.
- [Update Trace](https://docs.pandaprobe.com/api-reference/traces/update-trace.md): Partially update a trace.
- [Changelog](https://docs.pandaprobe.com/changelog/index.md): Release notes and version history for PandaProbe
- [Introduction](https://docs.pandaprobe.com/evaluation/agent-evaluation/introduction.md): Evaluate agent reliability and consistency across entire sessions.
- [Metrics](https://docs.pandaprobe.com/evaluation/agent-evaluation/metrics.md): Detailed reference for session-level agent evaluation metrics: reliability and consistency.
- [Concepts](https://docs.pandaprobe.com/evaluation/concepts.md): Core concepts behind PandaProbe's evaluation framework: runs, metrics, scores, signals, and monitors.
- [Overview](https://docs.pandaprobe.com/evaluation/overview.md): Evaluate the quality and reliability of your AI agents with trace and session-level metrics.
- [Introduction](https://docs.pandaprobe.com/evaluation/setup/introduction.md): Set up and run evaluations via the PandaProbe dashboard or API.
- [Run Evaluations via API](https://docs.pandaprobe.com/evaluation/setup/run-eval-api.md): Create and manage evaluation runs programmatically using the PandaProbe API.
- [Run Evaluations via UI](https://docs.pandaprobe.com/evaluation/setup/run-eval-ui.md): Create and manage evaluation runs through the PandaProbe dashboard.
- [Scheduling Evaluations](https://docs.pandaprobe.com/evaluation/setup/scheduling.md): Set up automated recurring evaluation monitors with custom cadences and filters.
- [Introduction](https://docs.pandaprobe.com/evaluation/trace-evaluation/introduction.md): Evaluate individual traces with LLM-as-judge metrics and embedding analysis.
- [Metrics](https://docs.pandaprobe.com/evaluation/trace-evaluation/metrics.md): Reference for all 9 built-in trace-level evaluation metrics.
- [Installation](https://docs.pandaprobe.com/get-started/installation.md): Install the PandaProbe Python SDK and configure your environment
- [Quickstart](https://docs.pandaprobe.com/get-started/quickstart.md): Trace your first LLM call in under 2 minutes
- [Welcome](https://docs.pandaprobe.com/get-started/welcome.md): Agent engineering platform for tracing, evaluation, and monitoring
- [Concepts](https://docs.pandaprobe.com/tracing/concepts.md): Traces, spans, span kinds, and the data model behind PandaProbe tracing
- [Conditional Tracing](https://docs.pandaprobe.com/tracing/configuration/conditional-tracing.md): Enable or disable tracing based on environment or runtime conditions
- [Environment Variables](https://docs.pandaprobe.com/tracing/configuration/environment-variables.md): Complete reference for all PandaProbe SDK environment variables
- [Project Configuration](https://docs.pandaprobe.com/tracing/configuration/project-configuration.md): Configure the SDK programmatically with pandaprobe.init()
- [Troubleshooting](https://docs.pandaprobe.com/tracing/configuration/troubleshooting.md): Common issues and solutions for PandaProbe SDK
- [Claude Agent SDK](https://docs.pandaprobe.com/tracing/integrations/claude-agent-sdk.md): Trace Claude Agent SDK conversations with automatic tool and thinking capture
- [CrewAI](https://docs.pandaprobe.com/tracing/integrations/crewai.md): Trace CrewAI crew executions with automatic agent and task tracking
- [Google ADK](https://docs.pandaprobe.com/tracing/integrations/google-adk.md): Trace Google Agent Development Kit agents with automatic lifecycle capture
- [LangGraph](https://docs.pandaprobe.com/tracing/integrations/langgraph.md): Trace LangGraph agent executions with automatic span capture
- [OpenAI Agents SDK](https://docs.pandaprobe.com/tracing/integrations/openai-agents-sdk.md): Trace OpenAI Agents SDK runs with first-class callback integration
- [Overview](https://docs.pandaprobe.com/tracing/integrations/overview.md): Automatic agent tracing for major agent development frameworks.
- [Context Managers](https://docs.pandaprobe.com/tracing/manual/context-managers.md): Fine-grained tracing with start_trace() and span() context managers
- [Decorators](https://docs.pandaprobe.com/tracing/manual/decorators.md): Trace functions automatically with `@trace` and `@span` decorators
- [Low-Level Client API](https://docs.pandaprobe.com/tracing/manual/low-level-client-api.md): Direct access to the PandaProbe Client for advanced use cases
- [Overview](https://docs.pandaprobe.com/tracing/overview.md): Understand how PandaProbe traces LLM applications across three layers of instrumentation
- [Programmatic Scoring](https://docs.pandaprobe.com/tracing/scoring.md): Attach programmatic scores to traces for evaluation and feedback tracking
- [Sessions](https://docs.pandaprobe.com/tracing/sessions.md): Group related traces into agent sessions for lifecycle-level analysis and evaluation
- [Users](https://docs.pandaprobe.com/tracing/users.md): Associate traces and sessions with end users, accounts, or tenants
- [Anthropic](https://docs.pandaprobe.com/tracing/wrappers/anthropic.md): Auto-trace Anthropic Messages API calls
- [Google Gemini](https://docs.pandaprobe.com/tracing/wrappers/google-gemini.md): Auto-trace Google Gemini generate_content API calls
- [OpenAI](https://docs.pandaprobe.com/tracing/wrappers/openai.md): Auto-trace OpenAI Chat Completions and Responses API calls
- [Overview](https://docs.pandaprobe.com/tracing/wrappers/overview.md): Zero-code LLM tracing for OpenAI, Anthropic, and Google Gemini

## OpenAPI Specs

- [openapi](https://docs.pandaprobe.com/openapi.json)