# HoneyHive AI Docs > AI observability and evaluation platform. Trace, evaluate, monitor, and improve AI agents and LLM applications. ## Documentation - [What is HoneyHive?](https://docs.honeyhive.ai/v2/introduction/what-is-hhai.md): HoneyHive is an AI observability and evaluation platform for tracing, testing, monitoring, and improving agents from development through production. ## Getting Started - [How to Trace Your First AI Application](https://docs.honeyhive.ai/v2/introduction/tracing-quickstart.md): In about 5 minutes, instrument an OpenAI call with HoneyHive, send your first trace, and inspect the full request and response in the dashboard. - [How to Run Your First HoneyHive Experiment](https://docs.honeyhive.ai/v2/introduction/experiments-quickstart.md): Run your first HoneyHive experiment in 5 minutes: define a function, score outputs with an evaluator, and compare results in the Experiments dashboard. - [Use with Coding Agents](https://docs.honeyhive.ai/v2/introduction/ai-coding-agents.md): Equip Cursor, Claude, and other AI coding agents with HoneyHive skills, CLI access, MCP docs search, and live SDK context for faster integrations. - [Troubleshooting & FAQs](https://docs.honeyhive.ai/v2/introduction/troubleshooting.md): Fix common HoneyHive tracing and evaluation issues: missing traces, export errors, session conflicts, SDK configuration problems, and dashboard FAQs. ## Observability - [AI Application Tracing with HoneyHive](https://docs.honeyhive.ai/v2/tracing/introduction.md): Start tracing AI applications with HoneyHive: capture LLM calls, tool invocations, and agent steps in a hierarchical execution view using OpenTelemetry. - [HoneyHive Tracing Concepts](https://docs.honeyhive.ai/v2/tracing/concepts.md): Learn HoneyHive's tracing data model: sessions, events, wide-event schema, and OpenTelemetry architecture for AI application observability. - [How to Create Custom Spans in HoneyHive](https://docs.honeyhive.ai/v2/tracing/custom-spans.md): Use the @trace decorator and context managers to trace business logic, workflows, and async functions beyond automatic LLM instrumentation. - [How to Initialize the HoneyHive Tracer](https://docs.honeyhive.ai/v2/tracing/tracer-initialization.md): Learn where to initialize HoneyHiveTracer for scripts, evaluate(), Lambda, and web servers, and how to create per-request session context. - [How to Enrich Traces with Metadata and Metrics](https://docs.honeyhive.ai/v2/tracing/enrich-traces.md): Add metadata, metrics, feedback, config, and user context to HoneyHive traces with enrich_span() and enrich_session() for filtering and evaluation. - [Schema Reference](https://docs.honeyhive.ai/v2/tracing/enrichment-schema.md): Reference HoneyHive enrichment namespaces, supported data types, and backend attribute names for metadata, metrics, feedback, config, and user properties. - [User Feedback](https://docs.honeyhive.ai/v2/tracing/setting-user-feedback.md): Capture thumbs-up/down ratings, comments, and implicit signals on HoneyHive traces to connect user sentiment with LLM outputs and agent behavior. - [Custom Metrics](https://docs.honeyhive.ai/v2/tracing/client-side-evals.md): Log evaluation scores, guardrail results, and custom numeric metrics from your application code onto HoneyHive traces for monitoring and analysis. - [Configuration](https://docs.honeyhive.ai/v2/tracing/configuration-details.md): Log prompt templates, model parameters, git context, and deployment config on HoneyHive traces to correlate outputs with the settings that produced them. - [User Properties](https://docs.honeyhive.ai/v2/tracing/setting-user-properties.md): Attach user IDs, account tiers, and tenant context to HoneyHive traces with user properties for per-user filtering, segmentation, and analysis. - [Online Experiments](https://docs.honeyhive.ai/v2/tracing/online-experimentation.md): Tag HoneyHive traces with experiment IDs and variant names to analyze A/B tests, compare prompt or model changes, and measure impact in production. - [How to Set Up Distributed Tracing](https://docs.honeyhive.ai/v2/tracing/distributed-tracing.md): Link traces across microservices and Lambda functions so LLM calls from every service appear in one HoneyHive session for end-to-end debugging. - [How to Trace MCP Servers and Clients](https://docs.honeyhive.ai/v2/tracing/mcp-tracing.md): Instrument Model Context Protocol servers and clients with HoneyHive using OpenInference auto-tracing or @trace decorators for production MCP apps. - [Multi-Instance Tracing](https://docs.honeyhive.ai/v2/tracing/multi-instance.md): Run multiple HoneyHiveTracer instances for multi-tenant apps, A/B tests, or per-agent routing so each instance exports to the right project and session. - [Multi-Provider Tracing](https://docs.honeyhive.ai/v2/tracing/multi-provider.md): Trace workflows that call OpenAI, Anthropic, Bedrock, or other providers in the same HoneyHive session with multi-provider instrumentation patterns. - [Multi-Modal Tracing](https://docs.honeyhive.ai/v2/tracing/multi-modal.md): Trace multi-modal AI pipelines in HoneyHive that process images, audio, video, and text so media handling steps appear alongside LLM spans. - [Multi-Threading (Python)](https://docs.honeyhive.ai/v2/tracing/multithreading.md): Propagate HoneyHive trace context across Python threads so multi-threaded agent and RAG workloads keep correct parent-child span relationships intact. - [Sampling](https://docs.honeyhive.ai/v2/tracing/sampling.md): Control trace volume in high-throughput AI apps with HoneyHive sampling strategies so you capture errors, priority users, and representative traffic. - [Span Filtering](https://docs.honeyhive.ai/v2/tracing/filtering.md): Drop noisy framework spans from HoneyHive traces with prefix-based span filtering rules so LLM, tool, and custom events stay readable at scale. - [Tracing via API](https://docs.honeyhive.ai/v2/tracing/manual-instrumentation.md): Log AI application events to HoneyHive via the REST API when you cannot use the Python SDK, for custom runtimes or language-specific integrations. - [Explore in UI](https://docs.honeyhive.ai/v2/tracing/ui-flows.md): Navigate the HoneyHive Traces page to debug sessions, inspect span trees, filter events, and curate production data for datasets and experiments. - [Tree View](https://docs.honeyhive.ai/v2/tracing/tree-view.md): Use HoneyHive Tree View to inspect parent-child span relationships, drill into LLM and tool events, and debug hierarchical AI agent execution. - [Timeline View](https://docs.honeyhive.ai/v2/tracing/timeline-view.md): Use HoneyHive Timeline View as a Gantt chart to find latency bottlenecks, compare span durations, and pinpoint slow LLM or tool calls in a session. - [Graph View](https://docs.honeyhive.ai/v2/tracing/graph-view.md): Explore HoneyHive Graph View to visualize agent execution as a directed graph, surface latency bottlenecks, and follow branching tool and model paths. - [Trajectory View](https://docs.honeyhive.ai/v2/tracing/trajectory-view.md): Visualize agent behavior in HoneyHive Trajectory View as a bubble chart to spot loops, stuck steps, outliers, and costly execution paths across steps. - [Thread View](https://docs.honeyhive.ai/v2/tracing/thread-view.md): Follow multi-turn conversations in HoneyHive Thread View as a chronological message thread across sub-agents, tools, and internal context handoffs. - [Session Aggregations](https://docs.honeyhive.ai/v2/tracing/aggregation-logic.md): Understand reserved HoneyHive session metadata fields that the platform automatically aggregates from child spans, including cost, tokens, and duration totals. - [Export Data](https://docs.honeyhive.ai/v2/tracing/query-data.md): Query and export HoneyHive trace data programmatically via the REST API or SDK to build custom reports, dashboards, and downstream analytics pipelines. - [Monitor AI Application Performance in HoneyHive](https://docs.honeyhive.ai/v2/monitoring/overview.md): Track cost, latency, token usage, and evaluator scores in HoneyHive's monitoring dashboard to detect failures and quality drift in production. - [Custom Charts](https://docs.honeyhive.ai/v2/monitoring/charts.md): Build custom HoneyHive charts with the Discover interface to query traces, plot cost and latency trends, and save views to your project dashboard. - [Online Evaluations](https://docs.honeyhive.ai/v2/monitoring/onlineevals.md): Configure HoneyHive online evaluations to score production traces automatically with code or LLM judges and surface quality drift on your dashboard. - [Overview](https://docs.honeyhive.ai/v2/monitoring/alerts/alerts_overview.md): Learn how HoneyHive alerts monitor LLM cost, latency, error rates, and evaluator scores, including alert types, states, and notification channels. - [Creating Alerts](https://docs.honeyhive.ai/v2/monitoring/alerts/alerts.md): Create HoneyHive alerts on AI performance, quality, and cost metrics with step-by-step rules so your team gets notified before users see failures. ## Evaluation - [AI Experiments and Evaluations in HoneyHive](https://docs.honeyhive.ai/v2/evaluation/introduction.md): Run offline experiments on datasets with evaluate() to compare prompts, models, and pipelines, score outputs, and catch regressions before deploy. - [Concepts](https://docs.honeyhive.ai/v2/evaluation/concepts.md): Understand how HoneyHive experiments are structured: functions, datasets, evaluators, runs, and trace linking for reproducible AI quality measurement. - [Comparing Experiments](https://docs.honeyhive.ai/v2/evaluation/comparing_evals.md): Compare HoneyHive experiment runs side by side to spot prompt regressions, model improvements, and metric deltas across versions before you ship changes. - [Sync Offline Evaluations](https://docs.honeyhive.ai/v2/evaluation/sync-offline-results.md): Push evaluation results you already computed into HoneyHive experiments - no server-side evaluators or SDK required - [Experiments via API](https://docs.honeyhive.ai/v2/evaluation/via-api.md): Run HoneyHive experiments through the REST API from TypeScript, Go, or CI pipelines when you need custom orchestration outside the Python evaluate() SDK. - [CI Regression Detection](https://docs.honeyhive.ai/v2/evaluation/ci-regression-detection.md): Gate pull requests on HoneyHive evaluation metrics with GitHub Actions so prompt, model, or pipeline changes fail CI when quality scores drop. - [Introduction](https://docs.honeyhive.ai/v2/evaluators/introduction.md): Overview of HoneyHive evaluators for measuring AI output quality offline and online. Learn evaluator types, when to use each, and how they fit your workflow. - [Client-Side Evaluators](https://docs.honeyhive.ai/v2/evaluators/client_side.md): Run client-side evaluators in your application code during experiments and CI. Compute scores locally with the SDK and send results to HoneyHive with traces. - [Python Evaluators](https://docs.honeyhive.ai/v2/evaluators/python.md): Create custom server-side Python evaluators in HoneyHive to score AI outputs with your own logic. Write deterministic checks and return structured metrics. - [LLM Evaluators](https://docs.honeyhive.ai/v2/evaluators/llm.md): Create LLM-powered evaluators in HoneyHive using custom prompts and rubrics. Score traces and experiment outputs with model-based judges and event filters. - [Use Portkey for LLM Evaluators](https://docs.honeyhive.ai/v2/evaluators/portkey.md): Route LLM evaluator calls through Portkey's AI gateway to access any of its 1,600+ supported models. - [Human Evaluators](https://docs.honeyhive.ai/v2/evaluators/human.md): Create human evaluator fields in HoneyHive for manual review and annotation of AI outputs. Collect reviewer scores, notes, and labels on traces and results. - [Composite Evaluators](https://docs.honeyhive.ai/v2/evaluators/composites.md): Combine multiple HoneyHive evaluators into one composite score with weighted aggregation. Roll up accuracy, safety, and style checks into one metric. - [Version Control](https://docs.honeyhive.ai/v2/evaluators/versioning.md): Version and roll back HoneyHive evaluators as prompts and logic change. Track evaluator history, compare versions, and restore prior definitions. - [Evaluator Template List](https://docs.honeyhive.ai/v2/evaluators/evaluator-templates.md): Browse HoneyHive server-side evaluator templates for common quality checks. Use built-in templates for toxicity, relevance, JSON validity, and more. - [Introduction](https://docs.honeyhive.ai/v2/datasets/introduction.md): Learn how HoneyHive datasets store test cases and ground truth for experiments, CI/CD evals, and fine-tuning. Centralize curated datapoints for AI development. - [Curate from Traces](https://docs.honeyhive.ai/v2/datasets/dataset-curation.md): Build HoneyHive datasets from production traces and sessions. Filter logged events in the UI or SDK, add ground truth, and curate regression sets. - [Upload Datasets](https://docs.honeyhive.ai/v2/datasets/import.md): Upload datasets to HoneyHive through the web UI or Python SDK. Import JSON, JSONL, and CSV files with inputs, ground truth, and metadata for evals. - [Run Experiments with HoneyHive Datasets](https://docs.honeyhive.ai/v2/datasets/run-experiments.md): Run experiments against HoneyHive datasets with evaluate(dataset_id=...). HoneyHive loads datapoints and aggregates run results server-side while your function and evaluators execute locally. - [Sync from External Sources](https://docs.honeyhive.ai/v2/datasets/sync.md): Keep HoneyHive datasets synced with S3, databases, and other external sources. Automate updates so experiment and evaluation data stays current. - [Export Datasets](https://docs.honeyhive.ai/v2/datasets/export.md): Export HoneyHive datasets programmatically with the Python SDK. Download curated datapoints as JSON for fine-tuning, external tools, or backup workflows. - [Import from Hugging Face](https://docs.honeyhive.ai/v2/datasets/hf-datasets.md): Import datasets from Hugging Face Datasets into HoneyHive for experiments and evaluation. Load HF splits via the SDK and map columns to datapoints. - [Annotation Queues](https://docs.honeyhive.ai/v2/evaluation/annotation-queues.md): Create HoneyHive annotation queues for human review of agent outputs, collect expert labels, and turn judgments into ground truth for evaluators. ## Prompt Management - [Managing Prompts](https://docs.honeyhive.ai/v2/prompts/overview.md): Create, test, version, and manage prompts in the HoneyHive Playground. Iterate on templates, compare models, and deploy prompt versions to your projects. - [Using Prompts in Code](https://docs.honeyhive.ai/v2/prompts/deploy.md): Learn to fetch deployed prompts from HoneyHive by name or version and integrate them in your application code using the Python SDK or TypeScript API SDK. ## Administration - [Organization Hierarchy](https://docs.honeyhive.ai/v2/workspace/organization-hierarchy.md): Understand how organizations, workspaces, and projects structure your HoneyHive account and control access to traces, datasets, evaluators, and provider keys. - [Managing Projects](https://docs.honeyhive.ai/v2/workspace/projects.md): Create and manage HoneyHive projects to organize apps, separate environments, and control access to traces, datasets, evaluators, and prompts per app. - [Inviting Teammates](https://docs.honeyhive.ai/v2/workspace/inviting-teammates.md): Learn how to invite teammates to your HoneyHive organization, workspaces, and projects, then assign roles so they can access traces, datasets, and evaluators. - [Role Based Access Control](https://docs.honeyhive.ai/v2/workspace/roles.md): Manage role-based access across your HoneyHive organization, workspaces, and projects. Control who can view traces, edit evaluators, and manage datasets. - [Provider Keys](https://docs.honeyhive.ai/v2/workspace/provider-keys.md): Configure AI provider API keys for LLM evaluators and the Playground in HoneyHive. Store credentials securely and scope provider access by workspace. - [Templates](https://docs.honeyhive.ai/v2/workspace/templates.md): Configure standard evaluators and monitoring charts as org-wide templates so every HoneyHive project inherits baseline evaluators, charts, and dashboards. - [Usage](https://docs.honeyhive.ai/v2/workspace/usage.md): View event counts, enrichment metrics, and usage trends in Organization Settings, then export reports to track HoneyHive usage across workspaces and projects. - [Multi-Tenant SaaS](https://docs.honeyhive.ai/v2/setup/managed.md): Get started on HoneyHive multi-tenant SaaS cloud. Create your account, connect projects, and start sending traces to HoneyHive's shared cloud in AWS US-West-2. - [Dedicated Cloud](https://docs.honeyhive.ai/v2/setup/dedicated.md): Get started on HoneyHive Dedicated Cloud with an isolated Data Plane in your AWS region. Meet data residency, compliance, and private networking requirements. - [Self-Hosting Overview](https://docs.honeyhive.ai/v2/setup/self-hosted.md): Deploy HoneyHive in your AWS environment. Plan architecture, install Helm charts, and configure a fully self-hosted HoneyHive platform. - [Security Architecture](https://docs.honeyhive.ai/v2/setup/self-hosted/security.md): Review security controls, encryption, authentication, and compliance requirements for HoneyHive self-hosted deployments in regulated or air-gapped environments. - [Data Flow & Residency](https://docs.honeyhive.ai/v2/setup/self-hosted/data-flow.md): Understand data classification, flow boundaries, and residency controls in HoneyHive self-hosted deployments. See where trace and evaluation data is stored. - [Operations Guide](https://docs.honeyhive.ai/v2/setup/self-hosted/operations.md): Use this guide after installing self-hosted HoneyHive to handle upgrades, monitoring, backups, scaling, incident response, and keep clusters reliable. - [Application Metrics](https://docs.honeyhive.ai/v2/setup/self-hosted/metrics.md): Use this reference to scrape per-service Prometheus metrics for signup, ingestion, trace storage, evaluation jobs, and the LLM proxy in self-hosted HoneyHive. - [Infrastructure Requirements](https://docs.honeyhive.ai/v2/setup/infrastructure-requirements.md): Review supported dependency versions, required Kubernetes operators, and infrastructure prerequisites before you deploy HoneyHive in a self-hosted environment. - [HoneyHive Platform Architecture](https://docs.honeyhive.ai/v2/platform-architecture.md): Understand how HoneyHive separates the Control Plane and Data Plane so traces, evaluations, and datasets stay isolated while you scale hosting models. - [Security](https://docs.honeyhive.ai/v2/setup/security.md): Learn how HoneyHive protects your data, infrastructure, and AI applications with encryption, access controls, network security, and compliance practices. ## Learn More - [HoneyHive Platform Concepts](https://docs.honeyhive.ai/v2/concepts.md): Learn core HoneyHive concepts: projects, sessions, events, datasets, experiments, evaluators, and prompts. See how they connect observability and evaluation. - [End-to-End: Multi-Agent Tracing and Evaluation](https://docs.honeyhive.ai/v2/tutorials/multi-agent-cookbook.md): Build a multi-agent customer support bot with Google ADK, add HoneyHive tracing, and run evaluations end to end. Follow a cookbook from agents to quality. - [Add Tracing to Existing Apps](https://docs.honeyhive.ai/v2/tutorials/add-tracing-5min.md): Add HoneyHive tracing to an existing LLM app in about five minutes. Install the SDK, initialize HoneyHiveTracer, instrument your provider, and see traces. - [Enrich Your Traces](https://docs.honeyhive.ai/v2/tutorials/enriching-traces.md): Enrich HoneyHive traces with user IDs, session metadata, and custom attributes. Make production spans easier to filter, debug, and evaluate with context. - [Deploy Tracing to Production](https://docs.honeyhive.ai/v2/tutorials/production-deployment.md): Deploy HoneyHive tracing to production with environment config, sampling, and reliability patterns. Set API keys, batching, and settings for LLM workloads. - [Trace Distributed Systems](https://docs.honeyhive.ai/v2/tutorials/distributed-tracing.md): Trace requests across services with HoneyHive context propagation. Connect parent and child spans over HTTP, queues, or RPC for multi-service LLM workflows. ## SDK Reference - [SDK Overview](https://docs.honeyhive.ai/v2/sdk-reference/overview.md): Compare HoneyHive Python and TypeScript SDKs to pick the right client for tracing, evaluation workflows, and REST API access in your stack. ## Python SDK - [Python SDK](https://docs.honeyhive.ai/v2/sdk-reference/python-sdk-ref.md): Explore the HoneyHive Python SDK reference for OpenTelemetry tracing, decorators, evaluations, and configuration options in your LLM application code. - [Environment Variables](https://docs.honeyhive.ai/v2/sdk-reference/environment-variables.md): Configure the HoneyHive Python SDK with environment variables for API keys, endpoints, tracer behavior, and sampling without hardcoding secrets in code. - [Python SDK Changelog](https://docs.honeyhive.ai/v2/sdk-reference/python-sdk-changelog.md): Track HoneyHive Python SDK releases, breaking changes, new features, and bug fixes so you can plan upgrades for your tracing and evaluation code. - [Migrate to SDK v1](https://docs.honeyhive.ai/v2/sdk-reference/python/migration/v0-to-v1.md): Upgrade HoneyHive Python SDK projects from v0.x to v1 with code rewrites for tracing, evaluations, and v2-ready instrumentation patterns. - [Migrate from Logger to v1](https://docs.honeyhive.ai/v2/sdk-reference/python/migration/logger-to-v1.md): Replace honeyhive-logger start, log, and update calls with HoneyHive Python SDK v1 tracer APIs so legacy instrumentation keeps sending traces. - [Migrate from honeyhive-bundled to v1](https://docs.honeyhive.ai/v2/sdk-reference/python/migration/honeyhive-bundled-to-v1.md): Switch from the honeyhive-bundled package to the main HoneyHive Python SDK v1 install with updated imports, config, and tracing setup steps. ## TypeScript API SDK - [TypeScript API SDK](https://docs.honeyhive.ai/v2/sdk-reference/typescript.md): Install and use the type-safe HoneyHive TypeScript API client to manage datasets, experiments, traces, and metrics with full editor autocompletion. - [Migrate from @honeyhive/logger](https://docs.honeyhive.ai/v2/sdk-reference/typescript-logger-to-api-sdk-migration.md): Migrate from @honeyhive/logger to @honeyhive/api-client with step-by-step rewrites for sessions, events, and configuration in TypeScript projects. - [Environment Variables](https://docs.honeyhive.ai/v2/sdk-reference/typescript/environment-variables.md): Configure the HoneyHive TypeScript API client with environment variables for API keys, data plane URLs, and request timeouts in Node or browser apps. - [Namespaces](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/namespaces.md): Browse HoneyHive TypeScript SDK namespaces for charts, datasets, events, experiments, metrics, and sessions, with links to every typed API method. - [Client](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/classes/Client.md): Initialize the HoneyHive TypeScript Client class with API keys and data plane URLs, then access typed namespaces for datasets, events, and experiments. - [FetchOptions](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/classes/FetchOptions.md): Per-request options for HoneyHive TypeScript SDK methods - [Charts Methods](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/charts/index.md): Call HoneyHive Charts API methods in TypeScript to create, list, update, and delete saved metric visualizations with typed request and response shapes. - [CreateChartRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/charts/CreateChartRequest.md): Use CreateChartRequest to define chart creation payloads with the HoneyHive Charts TypeScript client for typed calls and autocomplete. - [CreateChartResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/charts/CreateChartResponse.md): Read CreateChartResponse to parse chart creation results from the HoneyHive Charts TypeScript client with typed fields and autocomplete. - [DeleteChartRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/charts/DeleteChartRequest.md): Use DeleteChartRequest to define chart deletion payloads with the HoneyHive Charts TypeScript client for typed calls and autocomplete. - [DeleteChartResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/charts/DeleteChartResponse.md): Read DeleteChartResponse to parse chart deletion results from the HoneyHive Charts TypeScript client with typed fields and autocomplete. - [GetChartRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/charts/GetChartRequest.md): Use GetChartRequest to define single chart lookup payloads with the HoneyHive Charts TypeScript client for typed calls and autocomplete. - [GetChartResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/charts/GetChartResponse.md): Read GetChartResponse to parse single chart lookup results from the HoneyHive Charts TypeScript client with typed fields and autocomplete. - [GetChartsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/charts/GetChartsResponse.md): Read GetChartsResponse to parse chart list results from the HoneyHive Charts TypeScript client with typed fields and autocomplete. - [UpdateChartRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/charts/UpdateChartRequest.md): Use UpdateChartRequest to define chart update payloads with the HoneyHive Charts TypeScript client for typed calls and autocomplete. - [UpdateChartResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/charts/UpdateChartResponse.md): Read UpdateChartResponse to parse chart update results from the HoneyHive Charts TypeScript client with typed fields and autocomplete. - [Datapoints Methods](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/index.md): Use the Datapoints namespace in the HoneyHive TypeScript SDK to create, fetch, update, and batch-load dataset records with strongly typed API calls. - [BatchCreateDatapointsRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/BatchCreateDatapointsRequest.md): Use BatchCreateDatapointsRequest to define batch datapoint creation payloads with the HoneyHive Datapoints TypeScript client for typed calls and autocomplete. - [BatchCreateDatapointsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/BatchCreateDatapointsResponse.md): Read BatchCreateDatapointsResponse to parse batch datapoint creation results from the HoneyHive Datapoints TypeScript client with typed fields and autocomplete. - [CreateDatapointRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/CreateDatapointRequest.md): Use CreateDatapointRequest to define datapoint creation payloads with the HoneyHive Datapoints TypeScript client for typed calls and autocomplete. - [CreateDatapointResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/CreateDatapointResponse.md): Read CreateDatapointResponse to parse datapoint creation results from the HoneyHive Datapoints TypeScript client with typed fields and autocomplete. - [DeleteDatapointRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/DeleteDatapointRequest.md): Use DeleteDatapointRequest to define datapoint deletion payloads with the HoneyHive Datapoints TypeScript client for typed calls and autocomplete. - [DeleteDatapointResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/DeleteDatapointResponse.md): Read DeleteDatapointResponse to parse datapoint deletion results from the HoneyHive Datapoints TypeScript client with typed fields and autocomplete. - [GetDatapointRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/GetDatapointRequest.md): Use GetDatapointRequest to define single datapoint lookup payloads with the HoneyHive Datapoints TypeScript client for typed calls and autocomplete. - [GetDatapointResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/GetDatapointResponse.md): Read GetDatapointResponse to parse single datapoint lookup results from the HoneyHive Datapoints TypeScript client with typed fields and autocomplete. - [GetDatapointsRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/GetDatapointsRequest.md): Use GetDatapointsRequest to define datapoint list query payloads with the HoneyHive Datapoints TypeScript client for typed calls and autocomplete. - [GetDatapointsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/GetDatapointsResponse.md): Read GetDatapointsResponse to parse datapoint list results from the HoneyHive Datapoints TypeScript client with typed fields and autocomplete. - [UpdateDatapointRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/UpdateDatapointRequest.md): Use UpdateDatapointRequest to define datapoint update payloads with the HoneyHive Datapoints TypeScript client for typed calls and autocomplete. - [UpdateDatapointResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datapoints/UpdateDatapointResponse.md): Read UpdateDatapointResponse to parse datapoint update results from the HoneyHive Datapoints TypeScript client with typed fields and autocomplete. - [Datasets Methods](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/index.md): Manage evaluation datasets in TypeScript with the HoneyHive Datasets namespace for creating collections, adding datapoints, and listing test sets. - [AddDatapointsRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/AddDatapointsRequest.md): Use AddDatapointsRequest to define dataset datapoint addition payloads with the HoneyHive Datasets TypeScript client for typed calls and autocomplete. - [AddDatapointsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/AddDatapointsResponse.md): Read AddDatapointsResponse to parse dataset datapoint addition results from the HoneyHive Datasets TypeScript client with typed fields and autocomplete. - [CreateDatasetRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/CreateDatasetRequest.md): Use CreateDatasetRequest to define dataset creation payloads with the HoneyHive Datasets TypeScript client for typed calls and autocomplete. - [CreateDatasetResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/CreateDatasetResponse.md): Read CreateDatasetResponse to parse dataset creation results from the HoneyHive Datasets TypeScript client with typed fields and autocomplete. - [DeleteDatasetRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/DeleteDatasetRequest.md): Use DeleteDatasetRequest to define dataset deletion payloads with the HoneyHive Datasets TypeScript client for typed calls and autocomplete. - [DeleteDatasetResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/DeleteDatasetResponse.md): Read DeleteDatasetResponse to parse dataset deletion results from the HoneyHive Datasets TypeScript client with typed fields and autocomplete. - [GetDatasetsRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/GetDatasetsRequest.md): Use GetDatasetsRequest to define dataset list query payloads with the HoneyHive Datasets TypeScript client for typed calls and autocomplete. - [GetDatasetsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/GetDatasetsResponse.md): Read GetDatasetsResponse to parse dataset list results from the HoneyHive Datasets TypeScript client with typed fields and autocomplete. - [RemoveDatapointRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/RemoveDatapointRequest.md): Use RemoveDatapointRequest to define datapoint removal payloads with the HoneyHive Datasets TypeScript client for typed calls and autocomplete. - [RemoveDatapointResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/RemoveDatapointResponse.md): Read RemoveDatapointResponse to parse datapoint removal results from the HoneyHive Datasets TypeScript client with typed fields and autocomplete. - [UpdateDatasetRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/UpdateDatasetRequest.md): Use UpdateDatasetRequest to define dataset update payloads with the HoneyHive Datasets TypeScript client for typed calls and autocomplete. - [UpdateDatasetResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/datasets/UpdateDatasetResponse.md): Read UpdateDatasetResponse to parse dataset update results from the HoneyHive Datasets TypeScript client with typed fields and autocomplete. - [Events Methods](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/events/index.md): Read and write HoneyHive trace events through the TypeScript Events namespace with typed create, batch ingest, search, and update API methods. - [CreateEventBatchRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/events/CreateEventBatchRequest.md): Use CreateEventBatchRequest to define batch event ingestion payloads with the HoneyHive Events TypeScript client for typed calls and autocomplete. - [CreateEventBatchResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/events/CreateEventBatchResponse.md): Read CreateEventBatchResponse to parse batch event ingestion results from the HoneyHive Events TypeScript client with typed fields and autocomplete. - [CreateEventRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/events/CreateEventRequest.md): Use CreateEventRequest to define trace event creation payloads with the HoneyHive Events TypeScript client for typed calls and autocomplete. - [CreateEventResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/events/CreateEventResponse.md): Read CreateEventResponse to parse trace event creation results from the HoneyHive Events TypeScript client with typed fields and autocomplete. - [GetEventRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/events/GetEventRequest.md): TypeScript request type for the HoneyHive API - [GetEventResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/events/GetEventResponse.md): TypeScript response type from the HoneyHive API - [SearchEventsRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/events/SearchEventsRequest.md): Use SearchEventsRequest to define trace event search query payloads with the HoneyHive Events TypeScript client for typed calls and autocomplete. - [SearchEventsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/events/SearchEventsResponse.md): Read SearchEventsResponse to parse trace event search results from the HoneyHive Events TypeScript client with typed fields and autocomplete. - [UpdateEventRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/events/UpdateEventRequest.md): Use UpdateEventRequest to define trace event update payloads with the HoneyHive Events TypeScript client for typed calls and autocomplete. - [Experiments Methods](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/index.md): Run and compare HoneyHive evaluation experiments in TypeScript using typed methods for runs, summaries, metrics, schemas, and side-by-side analysis. - [CreateRunRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/CreateRunRequest.md): Use CreateRunRequest to define experiment run creation payloads with the HoneyHive Experiments TypeScript client for typed calls and autocomplete. - [CreateRunResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/CreateRunResponse.md): Read CreateRunResponse to parse experiment run creation results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [DeleteRunRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/DeleteRunRequest.md): Use DeleteRunRequest to define experiment run deletion payloads with the HoneyHive Experiments TypeScript client for typed calls and autocomplete. - [DeleteRunResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/DeleteRunResponse.md): Read DeleteRunResponse to parse experiment run deletion results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [GetExperimentCompareEventsRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetExperimentCompareEventsRequest.md): Use GetExperimentCompareEventsRequest to define event comparison query payloads via the HoneyHive Experiments TypeScript SDK with typed fields and autocomplete. - [GetExperimentCompareEventsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetExperimentCompareEventsResponse.md): Read GetExperimentCompareEventsResponse to parse event comparison results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [GetExperimentComparisonRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetExperimentComparisonRequest.md): Use GetExperimentComparisonRequest to build experiment comparison queries with the HoneyHive Experiments TypeScript client for typed calls and autocomplete. - [GetExperimentComparisonResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetExperimentComparisonResponse.md): Read GetExperimentComparisonResponse to parse experiment comparison results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [GetExperimentRunMetricsRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetExperimentRunMetricsRequest.md): Use GetExperimentRunMetricsRequest to define run metrics query payloads with the HoneyHive Experiments TypeScript client for typed calls and autocomplete. - [GetExperimentRunMetricsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetExperimentRunMetricsResponse.md): Read GetExperimentRunMetricsResponse to parse run metrics results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [GetExperimentSummaryRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetExperimentSummaryRequest.md): Use GetExperimentSummaryRequest to define experiment summary query payloads with the HoneyHive Experiments TypeScript client for typed calls and autocomplete. - [GetExperimentSummaryResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetExperimentSummaryResponse.md): Read GetExperimentSummaryResponse to parse experiment summary results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [GetRunRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetRunRequest.md): Use GetRunRequest to define single run lookup payloads with the HoneyHive Experiments TypeScript client for typed calls and autocomplete. - [GetRunResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetRunResponse.md): Read GetRunResponse to parse single run lookup results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [GetRunSchemaRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetRunSchemaRequest.md): Use GetRunSchemaRequest to define run schema query payloads with the HoneyHive Experiments TypeScript client for typed calls and autocomplete. - [GetRunSchemaResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetRunSchemaResponse.md): Read GetRunSchemaResponse to parse run schema results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [GetRunsRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetRunsRequest.md): Use GetRunsRequest to define run list query payloads with the HoneyHive Experiments TypeScript client for typed calls and autocomplete. - [GetRunsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetRunsResponse.md): Read GetRunsResponse to parse run list results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [GetRunsSchemaRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetRunsSchemaRequest.md): Use GetRunsSchemaRequest to define runs schema query payloads with the HoneyHive Experiments TypeScript client for typed calls and autocomplete. - [GetRunsSchemaResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/GetRunsSchemaResponse.md): Read GetRunsSchemaResponse to parse runs schema results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [UpdateRunRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/UpdateRunRequest.md): Use UpdateRunRequest to define experiment run update payloads with the HoneyHive Experiments TypeScript client for typed calls and autocomplete. - [UpdateRunResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/experiments/UpdateRunResponse.md): Read UpdateRunResponse to parse experiment run update results from the HoneyHive Experiments TypeScript client with typed fields and autocomplete. - [Metric Versions Methods](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metric-versions/index.md): Deploy and version HoneyHive evaluator metrics in TypeScript with the Metric Versions namespace for creating, listing, and promoting metric configs. - [CreateMetricVersionRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metric-versions/CreateMetricVersionRequest.md): Use CreateMetricVersionRequest to define metric version creation payloads with the HoneyHive MetricVersions TypeScript client for typed calls and autocomplete. - [CreateMetricVersionResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metric-versions/CreateMetricVersionResponse.md): Read CreateMetricVersionResponse to parse metric version creation from the HoneyHive MetricVersions TypeScript client with typed fields and autocomplete. - [DeployMetricVersionRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metric-versions/DeployMetricVersionRequest.md): Use DeployMetricVersionRequest to define metric version deployment payloads via the HoneyHive MetricVersions TypeScript SDK with typed fields and autocomplete. - [DeployMetricVersionResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metric-versions/DeployMetricVersionResponse.md): Read DeployMetricVersionResponse to parse metric version deployment from the HoneyHive MetricVersions TypeScript client with typed fields and autocomplete. - [GetMetricVersionsRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metric-versions/GetMetricVersionsRequest.md): Use GetMetricVersionsRequest to define metric version list query payloads with the HoneyHive MetricVersions TypeScript client for typed calls and autocomplete. - [GetMetricVersionsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metric-versions/GetMetricVersionsResponse.md): Read GetMetricVersionsResponse to parse metric version list results from the HoneyHive MetricVersions TypeScript client with typed fields and autocomplete. - [Metrics Methods](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/index.md): Create, run, and manage HoneyHive evaluator metrics in TypeScript with the Metrics namespace, including CRUD operations and on-demand metric execution. - [CreateMetricRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/CreateMetricRequest.md): Use CreateMetricRequest to define evaluator metric creation payloads with the HoneyHive Metrics TypeScript client for typed calls and autocomplete. - [CreateMetricResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/CreateMetricResponse.md): Read CreateMetricResponse to parse evaluator metric creation results from the HoneyHive Metrics TypeScript client with typed fields and autocomplete. - [DeleteMetricRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/DeleteMetricRequest.md): Use DeleteMetricRequest to define evaluator metric deletion payloads with the HoneyHive Metrics TypeScript client for typed calls and autocomplete. - [DeleteMetricResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/DeleteMetricResponse.md): Read DeleteMetricResponse to parse evaluator metric deletion results from the HoneyHive Metrics TypeScript client with typed fields and autocomplete. - [GetMetricsRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/GetMetricsRequest.md): Use GetMetricsRequest to define evaluator metric list query payloads with the HoneyHive Metrics TypeScript client for typed calls and autocomplete. - [GetMetricsResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/GetMetricsResponse.md): Read GetMetricsResponse to parse evaluator metric list results from the HoneyHive Metrics TypeScript client with typed fields and autocomplete. - [RunMetricRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/RunMetricRequest.md): Use RunMetricRequest to define on-demand metric execution payloads with the HoneyHive Metrics TypeScript client for typed calls and autocomplete. - [RunMetricResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/RunMetricResponse.md): Read RunMetricResponse to parse on-demand metric execution results from the HoneyHive Metrics TypeScript client with typed fields and autocomplete. - [UpdateMetricRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/UpdateMetricRequest.md): Use UpdateMetricRequest to define evaluator metric update payloads with the HoneyHive Metrics TypeScript client for typed calls and autocomplete. - [UpdateMetricResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/metrics/UpdateMetricResponse.md): Read UpdateMetricResponse to parse evaluator metric update results from the HoneyHive Metrics TypeScript client with typed fields and autocomplete. - [Sessions Methods](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/sessions/index.md): Start HoneyHive trace sessions in TypeScript and submit session event batches with the Sessions namespace for grouping spans in an invocation. - [CreateSessionEventBatchRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/sessions/CreateSessionEventBatchRequest.md): Use CreateSessionEventBatchRequest to define session event batch ingestion with the HoneyHive Sessions TypeScript client for typed calls and autocomplete. - [CreateSessionEventBatchResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/sessions/CreateSessionEventBatchResponse.md): Read CreateSessionEventBatchResponse to parse session event batch ingestion from the HoneyHive Sessions TypeScript client with typed fields and autocomplete. - [CreateSessionRequest](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/sessions/CreateSessionRequest.md): Use CreateSessionRequest to define trace session creation payloads with the HoneyHive Sessions TypeScript client for typed calls and autocomplete. - [CreateSessionResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/sessions/CreateSessionResponse.md): Read CreateSessionResponse to parse trace session creation results from the HoneyHive Sessions TypeScript client with typed fields and autocomplete. - [ApiError](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/classes/ApiError.md): Handle ApiError from the HoneyHive TypeScript SDK when API calls return non-success HTTP status codes, including status, message, and response details. - [ErrorResponse](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/classes/ErrorResponse.md): Inspect the ErrorResponse TypeScript type to read structured error payloads, status codes, and messages returned by failed HoneyHive API calls. - [HoneyHiveError](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/classes/HoneyHiveError.md): Extend or catch HoneyHiveError as the base error type for the HoneyHive TypeScript SDK when handling failed API calls and client-side failures. - [NetworkError](https://docs.honeyhive.ai/v2/sdk-reference/typescript/ref/classes/NetworkError.md): Catch NetworkError from the HoneyHive TypeScript SDK when requests fail due to connectivity, DNS, or transport issues before the API responds. ## Semantic Conventions - [Semantic Convention Reference](https://docs.honeyhive.ai/v2/sdk-reference/semconv-reference.md): Look up canonical HoneyHive trace attribute names, platform behaviors, and UI sideview mappings to structure spans and events for observability. - [Framework Attribute Mapping](https://docs.honeyhive.ai/v2/sdk-reference/semconv-alignment.md): Map OpenTelemetry GenAI, OpenInference, and Traceloop attributes to HoneyHive canonical keys so framework instrumentations normalize in your traces. ## OpenAPI Client - [OpenAPI Client (Any Language)](https://docs.honeyhive.ai/v2/sdk-reference/openapi-sdks.md): Generate HoneyHive REST API clients in any language from the public OpenAPI spec for serverless, polyglot, or regulated environments without vendor SDKs. ## CLI Reference - [Getting Started](https://docs.honeyhive.ai/v2/cli-reference/getting-started.md): Install and authenticate the HoneyHive CLI, then run your first commands to manage projects, traces, datasets, evaluators, and more from the terminal. - [Config as Code](https://docs.honeyhive.ai/v2/cli-reference/config-as-code.md): Learn to define HoneyHive evaluators, datasets, and platform resources as version-controlled files, then apply and sync them to your workspace with the CLI. - [Environment Variables](https://docs.honeyhive.ai/v2/cli-reference/environment-variables.md): Configure the HoneyHive CLI with environment variables. Reference supported variables for authentication, project scope, API endpoints, and runtime behavior. - [Namespaces](https://docs.honeyhive.ai/v2/cli-reference/namespaces.md): Explore HoneyHive CLI command namespaces. Find the right commands for sessions, events, datasets, metrics, experiments, and other platform resources. ## Reference - [Charts](https://docs.honeyhive.ai/v2/cli-reference/ref/charts.md): Define and manage saved charts. Charts are visualizations that aggregate metrics over time with bucketing, filters, and groupings. - [Datapoints](https://docs.honeyhive.ai/v2/cli-reference/ref/datapoints.md): Use the CLI to create, update, list, and delete datapoints in datasets, including batch imports and mapping individual records back to source trace events. - [Datasets](https://docs.honeyhive.ai/v2/cli-reference/ref/datasets.md): Use the HoneyHive CLI to create, list, update, and delete datasets. Organize datapoints into reusable test sets you can run during evaluations and experiments. - [Events](https://docs.honeyhive.ai/v2/cli-reference/ref/events.md): Use the CLI to read and write trace events. Events are spans that capture each step of your AI application's execution, inputs, outputs, and metadata. - [Experiments](https://docs.honeyhive.ai/v2/cli-reference/ref/experiments.md): Use the HoneyHive CLI to run, retrieve, and compare evaluation runs. Measure how prompt or configuration changes affect agent performance on datasets. - [Metric Versions](https://docs.honeyhive.ai/v2/cli-reference/ref/metric-versions.md): Use the CLI to snapshot, list, and deploy versions of a metric definition. Review changes and roll back evaluator configs without losing version history. - [Metrics](https://docs.honeyhive.ai/v2/cli-reference/ref/metrics.md): Define and run evaluators, i.e. automated quality checks that score traces against criteria like accuracy, safety, or correctness. - [Sessions](https://docs.honeyhive.ai/v2/cli-reference/ref/sessions.md): Use the CLI to group related trace events into sessions, the top-level container for multi-step or multi-service AI interactions and user conversations. ## API Reference - [API Reference (OpenAPI spec)](https://raw.githubusercontent.com/honeyhiveai/honeyhive-openapi/main/openapi.yaml): Machine-readable OpenAPI 3.0 spec for the HoneyHive REST API. ## Integrations - [Overview](https://docs.honeyhive.ai/v2/integrations/overview.md): Connect HoneyHive to the agent frameworks, model providers, and tools in your AI stack ## Agent Frameworks & Platforms - [How to integrate HoneyHive with Google ADK](https://docs.honeyhive.ai/v2/integrations/google-adk.md): Integrate HoneyHive with Google ADK to trace agent runs, tool calls, and model requests. Add four lines of setup with the OpenInference Google ADK instrumentor. - [How to trace AWS Strands Agents with HoneyHive](https://docs.honeyhive.ai/v2/integrations/strands.md): Trace AWS Strands Agents workflows, tool calls, and model requests with HoneyHive. Instrument Strands apps to capture multi-step agent execution in dashboard. - [How to trace Semantic Kernel with HoneyHive](https://docs.honeyhive.ai/v2/integrations/semantic-kernel.md): Trace Microsoft Semantic Kernel plugins, planners, and LLM calls with HoneyHive. Instrument your Semantic Kernel app to capture orchestration and model usage. - [How to trace AutoGen with HoneyHive](https://docs.honeyhive.ai/v2/integrations/autogen.md): Trace AutoGen AgentChat conversations, tool calls, and model requests with HoneyHive. Instrument multi-agent chats to capture delegation, tools, and LLM usage. - [How to trace LangGraph with HoneyHive](https://docs.honeyhive.ai/v2/integrations/langgraph.md): Trace LangGraph state graphs, node transitions, and LLM calls with HoneyHive. Add four lines of setup with the LangChain OpenInference instrumentor. - [How to trace CrewAI with HoneyHive](https://docs.honeyhive.ai/v2/integrations/crewai.md): Trace CrewAI crews, tasks, and agent collaboration with HoneyHive. Instrument crew workflows to capture role assignments, tool use, and model calls. - [How to trace OpenAI Agents SDK with HoneyHive](https://docs.honeyhive.ai/v2/integrations/openai-agents.md): Trace OpenAI Agents SDK runs, handoffs, and tool calls with HoneyHive. Instrument agent workflows to capture multi-step reasoning and model interactions. - [How to trace LangChain with HoneyHive](https://docs.honeyhive.ai/v2/integrations/langchain.md): Trace LangChain agents, chains, tools, and LLM calls with HoneyHive. Add four lines of setup and use OpenInference to instrument your existing LangChain app. - [How to trace LlamaIndex with HoneyHive](https://docs.honeyhive.ai/v2/integrations/llamaindex.md): Trace LlamaIndex query engines, retrieval, embeddings, and agents with HoneyHive. Add four lines of setup and use OpenInference to capture RAG workflows. - [How to trace Pydantic AI with HoneyHive](https://docs.honeyhive.ai/v2/integrations/pydantic-ai.md): Trace PydanticAI agents, tools, and LLM calls with HoneyHive. Add three lines of setup with HoneyHiveTracer.init() and Agent.instrument_all() for native OTel. - [How to trace Claude Agent SDK with HoneyHive](https://docs.honeyhive.ai/v2/integrations/claude-agent-sdk.md): Trace Claude Agent SDK sessions, tool use, and model calls with HoneyHive. Instrument agent runs to capture multi-turn workflows and execution details. - [How to trace Cursor SDK with HoneyHive](https://docs.honeyhive.ai/v2/integrations/cursor-sdk.md): Trace Cursor SDK agent runs, tool calls, and model requests with HoneyHive. Instrument Cursor SDK sessions to capture agent execution and debugging context. - [How to trace DSPy with HoneyHive](https://docs.honeyhive.ai/v2/integrations/dspy.md): Trace DSPy programs, modules, and LM calls with HoneyHive. Instrument DSPy pipelines to capture prompts, completions, and optimization runs in the dashboard. - [n8n](https://docs.honeyhive.ai/v2/integrations/n8n.md): Export n8n AI Agent workflow observability to HoneyHive with the execution-history shipper - [Microsoft Copilot Studio](https://docs.honeyhive.ai/v2/integrations/copilot-studio.md): Forward Microsoft Copilot Studio agent telemetry to HoneyHive with an Azure Function trace forwarder ## Model Providers - [How to trace OpenAI with HoneyHive](https://docs.honeyhive.ai/v2/integrations/openai.md): Trace OpenAI chat completions, tool calls, and embeddings with HoneyHive. Add four lines of setup and use OpenInference to capture every API call automatically. - [How to trace Anthropic with HoneyHive](https://docs.honeyhive.ai/v2/integrations/anthropic.md): Trace Anthropic Claude messages, tool use, and streaming with HoneyHive. Add four lines of setup and use OpenInference to capture every API call automatically. - [How to trace Gemini with HoneyHive](https://docs.honeyhive.ai/v2/integrations/gemini.md): Trace Google Gemini chat, embeddings, and multimodal requests with HoneyHive. Add four lines of setup and use OpenInference to capture Gemini API calls. - [How to trace Azure OpenAI with HoneyHive](https://docs.honeyhive.ai/v2/integrations/azure_openai.md): Trace Azure OpenAI chat completions, embeddings, and deployments with HoneyHive. Add four lines of setup and use OpenInference for Azure-hosted models. - [How to trace AWS Bedrock with HoneyHive](https://docs.honeyhive.ai/v2/integrations/aws_bedrock.md): Trace AWS Bedrock model invocations, Converse API calls, and streaming with HoneyHive. Add four lines of setup and use OpenInference for Bedrock automatically. ## AI Gateways - [How to trace LiteLLM with HoneyHive](https://docs.honeyhive.ai/v2/integrations/litellm.md): Trace LiteLLM proxy requests across providers with HoneyHive. Instrument LiteLLM to capture unified spans for chat completions, embeddings, and routed calls. - [How to integrate HoneyHive with Portkey](https://docs.honeyhive.ai/v2/integrations/portkey.md): Integrate HoneyHive with Portkey AI Gateway to trace routed LLM requests. Capture gateway spans, provider calls, and metadata from Portkey-enabled apps. - [How to integrate HoneyHive with TrueFoundry](https://docs.honeyhive.ai/v2/integrations/truefoundry.md): Integrate HoneyHive with TrueFoundry AI Gateway to export OTLP/HTTP traces. Route gateway spans to HoneyHive for unified observability across model providers. ## Coding Agents - [How to integrate HoneyHive with Claude Code](https://docs.honeyhive.ai/v2/integrations/claude-code.md): Integrate HoneyHive with Claude Code to export session traces for observability and evaluation. Send Claude Code runs to HoneyHive to debug agent quality. - [How to integrate HoneyHive with Devin](https://docs.honeyhive.ai/v2/integrations/devin.md): Integrate HoneyHive with Devin to export session traces for observability and evaluation. Send Devin runs to HoneyHive to review agent work and outcomes. ## Evaluation - [ASSERT](https://docs.honeyhive.ai/v2/integrations/assert.md): Run ASSERT behavior evaluations against HoneyHive-traced agents ## Changelog - [Product](https://docs.honeyhive.ai/v2/changelog/product.md): See what is new in the HoneyHive platform and UI. Browse product updates, feature launches, and major improvements that affect how you build and monitor AI. - [Clients](https://docs.honeyhive.ai/v2/changelog/clients.md): Browse release notes for the HoneyHive Python SDK, TypeScript API SDK, and CLI. Stay current with new features, bug fixes, and breaking API changes. - [Self-Hosting](https://docs.honeyhive.ai/v2/changelog/self-hosting.md): Review Helm chart updates and release notes for self-hosted HoneyHive. Find upgrade guidance, configuration changes, and migration steps for your deployment.