For data, research & AI teams
Newsletters carry analysis, launches and sponsorships that never reach a press release. NewsletterFIT turns each issue into records: the newsletter, the date, the sponsors, and the companies, people and topics mentioned. Pull them over a REST API or give an agent the MCP server.
7 days of Pro, no credit card required
The problem
They are spread across platforms, formats change per publisher, and sponsorships sit inside the body text. Scraping them yourself means maintaining parsers for every platform and every layout.
What you can do
Search newsletters, articles and entities and page through results with stable ids, from one API key.
REST APIThe MCP server works with Claude, ChatGPT, Cursor and other MCP clients, with OAuth sign-in or an API key.
MCP serverCompanies, products, people, guests, topics and sponsors are extracted per issue and linked to the newsletter they came from.
MethodologyEnterprise adds JSON exports of recent publications, articles, entities and the links between them.
ExportsExample
An example request and an abridged response. Every endpoint, parameter and limit is in the API reference.
curl "https://newsletterfit.com/api/v1/search?q=ai%20agents" \
-H "Authorization: Bearer nf_live_YOUR_KEY"{
"success": true,
"data": {
"query": "ai agents",
"newsletters": [
{
"id": 42,
"self": "/api/v1/publications/42",
"name": "The Ops Ledger",
"audience": { "value": 64000, "label": "64K", "basis": "substack" }
}
]
},
"meta": { "page": 1, "limit": 20, "total": 3 }
}How it works
01
Start the trial and create an API key from Billing.
02
REST for pipelines, MCP for agents and assistants.
03
API and MCP quotas grow with each plan; Enterprise adds exports.
FAQ
Newsletters, articles and entities (companies, products, people, guests, technologies, topics, brands and sponsors), with links between them. Article full text and raw files are not included.
Yes, an OpenAPI 3.1 file, plus a Markdown reference and MCP tool schemas for agents.
API and MCP calls have their own monthly quota per plan, separate from in-app usage, and a per-minute rate limit per key.
Contact support before using the data for model training or redistribution.