Skip to main content

​
What is llms.txt

llms.txt is an open standard created by llmstxt.org so that websites and APIs can expose a structured text file that language models can consume efficiently. It works like a robots.txt, but for LLMs: instead of saying what not to index, it says what is worth reading. The Timely.ai file contains:
  • A description of all REST endpoints (method, path, parameters, examples)
  • Webhook event types and payload structure
  • Authentication and rate limiting rules
  • Glossary of platform concepts (agents, channels, workspaces, etc.)
All of this in plain text, no HTML, no sidebar — ready to be injected as context into any LLM.

​
Where to find it

The file is available at the root of the documentation: 
https://docs.timelyai.com.br/llms.txt
Approximate size: ~75 KB. It fits within the context window of all modern models (Claude 3.x, GPT-4o, Gemini 1.5+).

​
How to download

curl https://docs.timelyai.com.br/llms.txt -o timely-llms.txt
To save directly to your project root: 
curl https://docs.timelyai.com.br/llms.txt -o ./timely-llms.txt

​
Use cases

When you add https://docs.timelyai.com.br as Docs in Cursor, the editor indexes the llms.txt automatically and makes it available via @Timely.ai. You don’t need to download the file manually.
Add a reference to the file in your CLAUDE.md so that Claude Code loads the context every time it starts a session in the project:
## Timely.ai API Context
Read the `timely-llms.txt` file at the project root to understand
the endpoints, authentication, and events of the Timely.ai API.
Then download the file once with curl and commit it to the repository, or configure a pre-session script to fetch the latest version.
In MCP servers that support context injection via local file, point to timely-llms.txt as the source document. The server injects the content into the system prompt before each model call.
Configure Continue.dev to index https://docs.timelyai.com.br as a context provider. It will read the llms.txt and the linked pages to build the semantic search index.

​
How the file is generated and updated

The llms.txt is generated automatically by the documentation build pipeline. Every time an .mdx page is modified and deployed, the file is regenerated and published at the same URL. To ensure you have the latest version in long-running projects, add a script to your workflow: 
#!/bin/bash
# scripts/update-timely-context.sh
curl -sL https://docs.timelyai.com.br/llms.txt -o ./timely-llms.txt
echo "llms.txt updated on $(date)"
You can run this script manually before intensive API development sessions, or schedule it via cron/CI.
The file contains no sensitive information — only public documentation. It is safe to commit it to your repository.