Biel.ai
Sign up
RAG API · Technical documentation

RAG API for technical documentation

Add source-linked documentation answers to your product through a REST API. Use Biel’s ingestion, retrieval, chat, feedback, and analytics endpoints without maintaining the indexing pipeline yourself.

Build with BielRead the API docs
OpenAPI 3.0Streaming responsesSource metadata
curlnodepython
# Ask any question of your indexed docs
curl https://app.biel.ai/api/v2/projects/acme-docs/chats/ \
  -H "Authorization: Api-Key $BIEL_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "message": "How do I rotate API keys?"
  }'

# Grounded response, with sources cited
{
  "answer": "Rotate keys from Settings → API
            Keys → Rotate...",
  "sources": [
    "keys/rotation.md",
    "security/best-practices.md"
  ],
  "chat_uuid": "9f4c…b2e1"
}

100,000+ technical questions answered, 98.7% rated helpful

ScyllaDB
Katalon
Tezos
Talon.one
GrepTime
CrazyGames
SecuroSys
ScyllaDB
Katalon
Tezos
Talon.one
GrepTime
CrazyGames
SecuroSys
The build vs. buy moment

What you would need to operate yourself

A documentation assistant needs ingestion, retrieval, model orchestration, source handling, authentication, monitoring, and ongoing updates.

Build it yourself

Build and maintain the pipeline

  • iIngest documentation, code blocks, and OpenAPI specifications.
  • iiOperate retrieval infrastructure and refresh the index after content changes.
  • iiiEvaluate prompts and models against representative documentation questions.
  • ivReturn source metadata and render citations in your application.
  • vAdd authentication, usage limits, monitoring, and cost controls.
  • viMaintain the pipeline as models, sources, and product requirements change.
The work continues as your sources, models, and product requirements change.
Use the Biel RAG API

Use Biel’s managed API

  • iConnect a supported documentation source and let Biel maintain the index.
  • iiUse REST endpoints documented with an OpenAPI specification.
  • iiiChoose streaming or buffered chat responses.
  • ivReceive source metadata with generated answers and retrieval results.
  • vAuthenticate requests and monitor usage through the Biel platform.
  • viAdopt model and retrieval updates without rebuilding the ingestion pipeline.
Start with the API documentation and a 14-day trial.
Endpoints

Endpoints for the documentation workflow

The API covers ingestion, sync, retrieval, conversational chat, feedback, and analytics. Use any HTTP client or generate one from the OpenAPI specification.

POST/v2/projects/<slug>/chats

Generate an answer

Send a question and get a grounded, source-linked answer. Streaming or buffered. Multi-turn aware: pass a chat_uuid to continue an existing conversation with full context.

GET/v2/projects/<slug>/search

Retrieve, don’t generate

Pure retrieval. Pass a query, get ranked chunks back with relevance scores and citations. For when you want to plug Biel into your own LLM stack and control generation yourself.

POST/v2/projects/<slug>/sources

Manage what’s indexed

Add a docs URL, a GitHub repo, an OpenAPI spec, a sitemap. Exclude paths, and trigger a recrawl any time through the sync endpoint to wire ingestion into CI/CD.

GET/v2/projects/<slug>/analytics/chat

Read conversation analytics

Every question, every answer, every cited source, with sibling reports for search, MCP, and insights. The same data the dashboard renders, available programmatically.

POST/v2/projects/<slug>/messages/<uuid>/feedback

Record feedback

Record thumbs-up, thumbs-down, or freeform feedback on any response. Surfaces in your dashboard, feeds back into ranking signals, helps you spot which sources actually answer well.

GETopenapi.json

Generate clients automatically

Full OpenAPI 3.0 specification. Use openapi-generator, Stainless, or your favorite SDK builder to scaffold typed clients in any language your team writes in.

What’s included

What Biel manages for you

i.

Streaming or buffered responses

Stream chat responses with server-sent events or request a buffered response for jobs that do not need incremental output.

ii.

API-key authentication

Authenticate API requests with project credentials that can be rotated from the dashboard.

iii.

Usage limits by plan

Published plans include defined interaction limits, with higher capacity available for larger deployments.

iv.

Sources in the response

Chat and retrieval responses include source information that your interface can display for verification.

v.

Index refresh controls

Refresh documentation from the dashboard, a schedule, or supported automation hooks.

vi.

Separate projects

Keep documentation sources and API usage organized in separate Biel projects under one account.

Use cases

What teams build with the API

Custom in-product chat

An assistant inside your app

Build the chat experience that fits your product. The widget is a starting point; the API lets you control everything: layout, persona, scope, escalation paths, follow-up actions, billable events.

Internal copilots

For sales, success, and support

Power internal tools with grounded answers from your knowledge base. Let your team query past tickets, runbooks, deal notes, and product docs from one prompt.

Pipelines and automations

Wire docs into your workflows

Auto-respond to GitHub issues with linked docs. Pre-populate Zendesk replies. Generate context-aware tooltips at build time. Anywhere your docs are useful, the API is.

“It consistently delivers accurate results, and we’ve had no issues depending on it as part of our documentation experience.”

Lars Wilhelmer · Documentation Engineer at Talon.One

Things teams ask before integrating

Which AI models do you support?

Biel supports selected OpenAI and Anthropic models. Model availability depends on the plan; the API documentation defines the stable request and response contract.

Can I bring my own LLM?

Yes, on Enterprise plans. Use the /v2/projects/<slug>/search endpoint to retrieve grounded chunks and feed them into your own model. You get the indexing and ranking infrastructure, you keep control of generation.

Is the API the right entry point for my team?

Use the API when you are building your own interface or workflow. For a finished experience, start with AI chat, AI search, or the hosted MCP server.

What about latency?

Latency depends on the selected model, question, and retrieved context. Streaming lets your interface display the response as it is generated.

Is anything used for training?

No. We do not use your indexed content or user conversations to train AI models. See the Security Overview.

Build with your docs through the API

Build your assistantRead the API docs
14-day free trial · No credit card required
Try me ↓