AnswerLine
Get AI answers, citations and source URLs from ChatGPT, Gemini, Copilot, Google Search, News, AI Overview and AI Mode for any prompt.
Hosted MCP Server
npx add-mcp 'https://mcp.answerline.dev/mcp'Installs into Claude Code, Codex, Cursor and more
Documentation
Documentation
Start with the quickstart, then read the guide for the part you are building. Every field and error code is in the API reference.
The guides below are ordered the way an integration usually goes. The quickstart gets a key and a first answer; authentication and errors covers the error envelope and which failures are safe to retry; rate limits and credits explain what your plan allows and what each request costs; async tasks and webhooks are the pattern for volume, where work is queued and results are delivered to you; MCP and integrations connect the API to agents and automation tools. The API reference is the full contract, every endpoint, field and error code.
Guides
Quickstart
From an API key to a synchronous answer, an async task and a webhook, with curl, the TypeScript SDK and the Python SDK.
Authentication and errors
API keys, the error envelope and every error code, which requests are safe to retry, and request ids.
Accounts, teams and API keys
Sign-in, owner and member roles, invites, creating and rotating API keys, the audit log, and exporting or deleting your data.
Rate limits and concurrency
The one-second rate window, concurrency slots per plan, queue capacity and the headers that report them.
Credits and billing
How requests are priced, reserved and charged, how billing cycles reset, and what happens when credits run out.
Async tasks and batches
Queue single tasks or batches of up to 500 with priorities and idempotency keys, collect results and manage the queue.
Webhooks
Receive task results by webhook: payload, signature verification, secret rotation, retries and delivery order.
Measuring visibility
Why one answer misleads, how many samples to take, and how to measure a brand's mention rate, rank and citations with plain HTTP or the SDK helpers.
MCP server
Connect AI agents and MCP clients to the hosted MCP server: endpoints, authentication, tools and costs.
Integrations
Call the API from n8n, Zapier and LangChain for Python and JavaScript.
Coverage and limits
Which engines are sold, the countries each answers from, US-state targeting, and which targeting fields each takes.
Engine features
Which include flags and result fields each served engine supports, what they cost, which come back empty today, and each engine's targeting.
Output formats
When to use text or include.markdown, which engine returns which, what each costs, and why include.html is refused today.
Versioning and deprecation
How the API is versioned, what counts as a breaking change, and how deprecations are announced.
API reference
Every endpoint, parameter and response code, generated from the OpenAPI document.
Engines
Each engine page lists its request options, the fields it returns and its price.