llms.txt
Compare with other sites' llms.txt files →# HoneyHive
> HoneyHive is an AI agent observability and evaluation platform. Teams
> instrument their agents and LLM applications with OpenTelemetry, trace
> what the agent actually did, evaluate every output, find the failure,
> and improve quality with each release. Used by enterprises running
> agents in production, including in regulated industries.
HoneyHive covers one loop: observe, evaluate, improve. Observability
without evaluation tells you an agent ran. Evaluation without
observability tells you a score with no way to find the cause. HoneyHive
connects the two so a failing output leads back to the span that caused it.
Deployment: cloud, or self-hosted for teams that cannot send trace data
to a vendor. Free up to 10,000 events per month.
## Product
- [Homepage](https://www.honeyhive.ai/): What HoneyHive does and who uses it.
- [Enterprise](https://www.honeyhive.ai/enterprise): Governance, self-hosted deployment, RBAC, audit-ready compliance (SOC 2 Type II, GDPR, HIPAA).
- [Pricing](https://www.honeyhive.ai/pricing): Plans and limits. Free tier is 10,000 events per month.
- [About](https://www.honeyhive.ai/about): Company, team, and mission.
- [Book a demo](https://www.honeyhive.ai/book-a-demo): Talk to the team.
## Documentation
Full docs have their own llms.txt with per-page links. Prefer these for
technical questions about how HoneyHive works.
- [Docs llms.txt](https://docs.honeyhive.ai/llms.txt): Structured index of all HoneyHive documentation.
- [Docs llms-full.txt](https://docs.honeyhive.ai/llms-full.txt): Full documentation text in one file.
- [What is HoneyHive?](https://docs.honeyhive.ai/v2/introduction/what-is-hhai): Platform overview.
- [Tracing quickstart](https://docs.honeyhive.ai/v2/introduction/tracing-quickstart): Instrument an OpenAI call and see a trace in about 5 minutes.
- [Experiments quickstart](https://docs.honeyhive.ai/v2/introduction/experiments-quickstart): Run an evaluation with a scorer and compare results.
- [Tracing concepts](https://docs.honeyhive.ai/v2/tracing/concepts): Sessions, events, the wide-event schema, and the OpenTelemetry architecture.
- [Evaluation](https://docs.honeyhive.ai/v2/evaluation/introduction): Offline and online evaluation.
- [Evaluators](https://docs.honeyhive.ai/v2/evaluators/introduction): LLM-as-judge, code-based, and human evaluators.
- [Monitoring and alerts](https://docs.honeyhive.ai/v2/monitoring/overview): Production monitoring and alerting on agent behavior.
- [Use with coding agents](https://docs.honeyhive.ai/v2/introduction/ai-coding-agents): Installable HoneyHive skills for Claude Code, Cursor, and other coding agents, so an agent can instrument a codebase from a prompt.
- [SDK reference](https://docs.honeyhive.ai/v2/sdk-reference/overview): Python and TypeScript SDKs, OpenTelemetry-native.
- [Troubleshooting and FAQs](https://docs.honeyhive.ai/v2/introduction/troubleshooting): Missing traces, export errors, session conflicts, SDK configuration.
## Integrations
Model providers: OpenAI, Anthropic, Google Gemini.
Frameworks: LangChain, LlamaIndex, and any OpenTelemetry-instrumented stack.
Coding agents: Claude Code, Cursor, Devin, GitHub Copilot.
Third-party agent platforms: ServiceNow, Microsoft Copilot Studio, Salesforce Agentforce.
## Writing
- [Blog](https://www.honeyhive.ai/blog): Technical posts on agent evaluation, observability, and the agent development lifecycle.
- [Changelog](https://docs.honeyhive.ai/v2/changelog/product): Product updates.
## Optional
- [Careers](https://careers.honeyhive.ai/)
- [Discord](https://discord.gg/vqctGpqA97)
- [LinkedIn](https://www.linkedin.com/company/honeyhiveai/)
- [X](https://x.com/honeyhiveai)