agent-data
Start free
Catalog / Cognition

Cognition API

cognition.ai

Lists articles exposed by Cognition's live blog index in source order, with cursors for older entries. Retrieves an individual article by its source slug or canonical URL, including publication and update dates, authors, categories, summary, and full public text when present. The API reads the source live and does not maintain a stored archive.

  • Business
  • cognition
  • blog
  • articles
  • software-engineering
Connect your agent Call over HTTP
3 endpointsDocs updated

Endpoint health

Loading verification…

Connect your agent

Connect through MCP so your agent can find this API, read its docs, and call its endpoints.

Setup guide

Add the server from your terminal.

Claude Code · terminal
claude mcp add --transport http agent-data https://agent-data.motie.dev/mcp

In Claude Code, run /mcp, select agent-data, then Authenticate. Sign in to your agent-data account in the browser.

Then ask your agent to use the Cognition API for your task.

Call it over HTTP

Get an API key

Sign up free, generate an API key, and set it as AGENT_DATA_API_KEY in your environment. Your first 100 calls each month are free.

Reads the current source-rendered article index, newest first. Pass the opaque `pagination.next_cursor` back as `cursor` to continue through the live source order; no stored archive is maintained.

Optional parameters 2

Maximum number of articles to return (1–50).

Opaque next_cursor token from the previous response; results continue with older source entries.

View call pricing
cURL · Articles
curl --request GET --get 'https://api.agent-data.dev/cognition-ai/v1/articles' \
  --header "Authorization: Bearer $AGENT_DATA_API_KEY"

Endpoints

Inputs and response fields for every endpoint in this API.

GETArticles/v1/articles

Reads the current source-rendered article index, newest first. Pass the opaque `pagination.next_cursor` back as `cursor` to continue through the live source order; no stored archive is maintained.

Input

limitOptionalQuery · integer, min: 1, max: 50, default: 20

Maximum number of articles to return (1–50).

Example: 5

cursorOptionalQuery · string | null

Opaque next_cursor token from the previous response; results continue with older source entries.

Response

dataobject
data.itemsobject[]
data.items[].titlestring
data.items[].summarystring | null
data.items[].article_idstring
data.items[].canonical_urlstring
data.items[].published_datestring | null
data.source_urlstring
data.source_transportstring
paginationobject
pagination.has_moreboolean
pagination.next_cursorstring | null
request_idstring

For the next page, pass next_cursor from the response as cursor. Leave it empty on the first call.

GETArticles by url/v1/articles/by-url

Fetches the live public article named by a cognition.com or cognition.ai blog URL.

Input

urlRequiredQuery · string

Canonical or cognition.ai alias URL for one public Cognition blog article.

Example: https://cognition.com/blog/devin-comes-to-sao-paulo

Response

dataobject
data.titlestring
data.authorsstring[]
data.summarystring | null
data.full_textstring | null
data.article_idstring
data.categoriesstring[]
data.updated_datestring | null
data.canonical_urlstring
data.published_datestring | null
data.source_transportstring
request_idstring
GETArticles by article id/v1/articles/{article_id}

Fetches an individual live Cognition blog article by its source path slug.

Input

article_idRequiredPath · string

Source slug from the article's Cognition blog URL.

Example: devin-comes-to-sao-paulo

Response

dataobject
data.titlestring
data.authorsstring[]
data.summarystring | null
data.full_textstring | null
data.article_idstring
data.categoriesstring[]
data.updated_datestring | null
data.canonical_urlstring
data.published_datestring | null
data.source_transportstring
request_idstring

Don’t see the endpoint you need?

Ask agent-data to add an endpoint or return more fields. Extensions are free and keep existing endpoints working.