Inkeep
AI agents and cited docs search for support and customer teams
Overview
Inkeep, a Y Combinator company, builds AI agents for customer-facing teams: cited answers over documentation and support content, plus an agent framework with a no-code visual builder and a TypeScript SDK in two-way sync. The framework is free to self-host under the Elastic License 2.0; the managed platform is sold by quote.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 14 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- Visual builder and TypeScript SDK
- MCP servers and tools
- Observability and UI library
- Deploy to Vercel or Docker
- Dedicated forward-deployed engineer
- Unified AI search (managed RAG)
- Use from Slack and support platforms
- PII removal and data controls
Use cases
What you can produce with Inkeep
- A support assistant that answers from your documentation and content with cited sources
- Agents built in a no-code visual builder or a TypeScript SDK, kept in two-way sync
- Multi-agent workflows triggered by webhooks, schedules or service events
- An OpenAI-compatible chat API for embedding answers in your own interface
- Agents that act through MCP tools, such as opening docs pull requests or tickets
- Slack and support-platform assistants for CX teams
Inkeep is a good fit for developer-tool and API companies with real documentation and ticket volume. Answers cite their sources, an OpenAI-compatible API makes custom front ends easy, and the agent framework lets engineers and support staff edit the same agent in code or in a visual builder.
Two things to know: the framework is source-available under the Elastic License, not open source, and the managed platform has no public price or free tier, only a 30-day sandbox after a sales call. Budgeting needs a conversation, and answer quality depends on the docs you feed it.
What we found
Inkeep is a San Francisco company from Y Combinator's Winter 2023 batch that builds AI agents for customer-facing teams. It started with AI search and chat over technical documentation, and that product still exists: an assistant that answers from a company's docs and other content, cites its sources and hands off to support channels, reachable through chat components, a Slack bot or an API documented as compatible with OpenAI's Chat Completions format. Its homepage names Anthropic, Midjourney, Pinecone, PostHog, Postman, Clay and Solana among its users. In its case studies Fingerprint is credited with cutting support tickets by 48%, and Payabli's documentation team uses query data to find gaps in its docs; both are Inkeep's own write-ups.
The bigger change is the agent platform. In September 2025 Inkeep announced a $13 million seed round led by Khosla Ventures, GreatPoint Ventures and Y Combinator, and published Inkeep Agents on GitHub: a no-code visual builder and a TypeScript SDK kept in two-way sync, so the same agent can be edited in either. Agents delegate to sub-agents, use MCP tools and fire on webhooks or schedules; Inkeep's showcase is an agent that turns a resolved support ticket into a pull request against the docs. Earlier versions of this page called the framework open source, as Inkeep does, but the repository's licence is the Elastic License 2.0 with supplemental terms. That allows free self-hosting while forbidding resale as a hosted service, which makes it source-available rather than open source.
Pricing follows the split. The self-hosted framework is free and deploys to Vercel or Docker. The managed platform has no free tier and no public price: Inkeep's FAQ offers a free sandbox built on a prospect's public content, available for 30 days after a sales call, followed by a scoped pilot. Enterprise adds a forward-deployed engineer, managed search, Slack and support-platform access, and PII controls. Price estimates on third-party sites are not Inkeep's and are not repeated here.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
About this listing
This entry was compiled from publicly available data including Inkeep's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Inkeep unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Inkeep directly →
Spotted something out of date? Suggest an update →
Alternatives to Inkeep
More in Business & Finance
