llms.txt

Give an AI assistant the xfinlink reference as plain text so the code it writes uses the right calls.

xfinlink publishes plain-text files written for AI assistants. Give one to Claude, ChatGPT, Cursor or another assistant before you ask it to write code against xfinlink. The files need no key.

The files

Ten files are published. Each one is served at the same path on https://xfinlink.com and on https://api.xfinlink.com, as UTF-8 plain text. https://www.xfinlink.com redirects to https://xfinlink.com. Each file holds whole pages of these docs, as Markdown.

https://xfinlink.com/llms.txt: the main file. It holds Quickstart and Python client reference, and ends with links to every docs page and to the eight endpoint files below.
https://xfinlink.com/llms-prices.txt: /v1/prices/{ticker}. Prices and Price fields.
https://xfinlink.com/llms-fundamentals.txt: /v1/fundamentals/{ticker}. Fundamentals and Fundamentals fields.
https://xfinlink.com/llms-metrics.txt: /v1/metrics/{ticker}. Metrics and Metrics catalogue.
https://xfinlink.com/llms-resolve.txt: /v1/resolve/{ticker}. Resolve.
https://xfinlink.com/llms-search.txt: /v1/search. Search.
https://xfinlink.com/llms-index.txt: /v1/index/{index_name} and /v1/index/{index_name}/events. Index membership and Index events.
https://xfinlink.com/llms-insiders.txt: /v1/insiders/{ticker}. Insider transactions and Insider and holdings fields.
https://xfinlink.com/llms-holdings.txt: /v1/holdings/{ticker}, /v1/managers and /v1/managers/{manager_id}/holdings. Institutional holdings, Manager search, Manager holdings and Insider and holdings fields.
https://xfinlink.com/llms-full.txt: every page of these docs, in sidebar order.

Responses carry Cache-Control: public, max-age=3600, so a cached copy can be up to one hour old.

Terminal

$curl -s https://xfinlink.com/llms.txt | head -3

stdout

# xfinlink

Python client for US equity data: prices, financial statements, computed metrics, entity resolution, search, index constituents and membership changes, insider transactions, and institutional (Form 13F) holdings. The same data is served over HTTPS at https://api.xfinlink.com and through the MCP server at https://api.xfinlink.com/mcp. ...

How `llms.txt` is organised

In order:

# xfinlink and a one-paragraph summary.
The Quickstart page, from its # Quickstart heading.
The Python client reference page, from its # Python client reference heading.
## Docs pages as Markdown: the Markdown address of every docs page, such as https://xfinlink.com/docs/prices.md.
## Endpoint files: the addresses of the eight endpoint files.

Each page starts with its title as a # heading, its one-line summary, its endpoint where it has one, and a Source: line with the page's address. Links between pages are full https://xfinlink.com/docs/... addresses. Code samples keep the YOUR_API_KEY placeholder.

How the endpoint files are organised

Each endpoint file opens with a heading that names its endpoints, such as # xfinlink: GET /v1/prices/{ticker}, and a line listing the pages it holds. The pages follow in the order listed above, each in the same form as in llms.txt.

Use it with an AI assistant

Pick the file that matches the task.

For Python work across several datasets, give the assistant llms.txt.
For REST work on one endpoint, give it that endpoint's file. It holds the full reference pages for those endpoints: parameters, fields, errors and edge cases.

Then give the file to the assistant in one of these ways.

Paste the text into the conversation, or add it to the project's knowledge or rules files, before you describe what you want.
Give the assistant the URL, if it can read web pages.
Load it from Python with xfl.llm_context(), below.

To let an assistant fetch live data itself instead of writing code, connect it to the MCP server. See MCP server and Vibe coding.

Load it from Python

xfl.llm_context() downloads llms.txt from https://api.xfinlink.com/llms.txt, prints it, and returns it as a str. It needs no key. If the pyperclip package is installed, it also copies the text to your clipboard.

request.py

import xfinlink as xfl

text = xfl.llm_context()