The Acenda API reference is published for AI tools at https://api.acenda.io/v1/docs/llms.txt, a plain-text link you can give to an AI coding tool or agent so it learns how to integrate with Acenda before it writes any code.

The link is a short index that points to the full plain-text reference (llms-full.txt), which covers authentication, request conventions, errors and the public endpoints. People reading the reference themselves can use the Acenda API documentation. Postman users can import the Postman collection.

Give the reference to any AI tool

Copy the link https://api.acenda.io/v1/docs/llms.txt.
Paste the link into your AI tool and ask it to read the reference before writing any code.
Describe what you want to build, such as a script, report or sync.
Check the code against the Acenda API documentation before you run it against live data.

If the tool can't open links, download llms-full.txt and attach the file to the chat, or paste its contents.

Example prompt

Copy this prompt and adjust the task to your needs:

Read https://api.acenda.io/v1/docs/llms.txt, then write a
Python script that pulls yesterday's orders from the Acenda
API using my Client ID and Client Secret from environment
variables.

Every request the tool writes needs two headers: Authorization: Bearer followed by an access token, and X-Astur-Organization set to your organization name. The reference explains how to get the token with your Client ID and Client Secret.

Important: Never paste your Client Secret into an AI chat. Keep it in environment variables or a secrets manager, and have the code read it from there. Create a separate API key for each tool or contractor so you can revoke one without affecting the others. A Client Secret is shown only once, when you create or regenerate the key under ADMINISTRATION > Users > API Keys.

ChatGPT

For a single task, paste the link or attach llms-full.txt in a new chat.
For ongoing work, select New project in the sidebar, add llms-full.txt to the project files, and add an instruction such as "Use the Acenda API reference in the project files" in the project settings.

Claude

For a single task, paste the link or attach llms-full.txt in a new chat.
For ongoing work, create a project, upload llms-full.txt to the project knowledge, and set project instructions that tell Claude to use the reference for Acenda API code.

Gemini

For a single task, paste the link or attach llms-full.txt in a new chat.
For ongoing work, go to Gems > New Gem, enter a name and instructions, click Add files under Knowledge to upload llms-full.txt, then click Save.

Grok

For a single task, paste the link in a chat, or click the + icon next to the message box to attach llms-full.txt.
For ongoing work, add llms-full.txt to a Grok project so every chat in the project can use the reference.

Coding agents and IDEs

Coding agents such as Cursor, Claude Code and GitHub Copilot read an instructions file in your project. Add the Acenda API rules to that file so the agent follows them in every session:

Cursor: add the rules to an AGENTS.md file in the project root, or create a project rule in the .cursor/rules folder.
Claude Code: add the rules to a CLAUDE.md file in the project root. Run /init to create one if the project has none.
GitHub Copilot: add the rules to .github/copilot-instructions.md or to an AGENTS.md file.

Example rules to paste into the file:

## Acenda API
- Before writing code that calls the Acenda API, read
  https://api.acenda.io/v1/docs/llms.txt and the full
  reference it links to.
- Use only endpoints, fields and parameters documented
  there. If something is not documented, say so.
- Read credentials from the ACENDA_CLIENT_ID,
  ACENDA_CLIENT_SECRET and ACENDA_ORGANIZATION environment
  variables. Never hard-code them.

For offline work, save llms-full.txt in the project and point the rules at the local copy.

Tip: To ask questions about your Acenda data without writing code, connect the Acenda MCP server instead. See How to connect your AI assistant to Acenda.

Troubleshooting & FAQ

What if my AI tool can't open links?

Download llms-full.txt and attach it to the chat, or paste its contents. Some tools only read links when web search or browsing is turned on.

Why does the AI tool invent endpoints or fields?

Tell the tool to use only the endpoints and fields in llms.txt or the Acenda API documentation, and to say so when something isn't documented. Check any endpoint you don't recognize against the documentation before you run the code.

Why did a request fail with 401 Unauthorized?

A 401 means the access token is missing, invalid or expired. Make sure the code requests a fresh token and sends it in the Authorization header, and that the environment variables hold the right values. If the Client Secret was lost or exposed, regenerate the key under ADMINISTRATION > Users > API Keys and update your environment variables.

Why did a request fail with 403 Forbidden?

A 403 means the API key doesn't have permission for that request, or the X-Astur-Organization header doesn't match your organization. Check the header value and the permissions assigned to the key.

Still need help? Contact Acenda Support.

Keywords

llms.txt, llms-full.txt, Acenda API reference for AI, AI coding tool, AI agent, coding agent, ChatGPT, Claude, Gemini, Grok, Cursor, Claude Code, GitHub Copilot, AGENTS.md, CLAUDE.md, copilot-instructions.md, Client Secret, environment variables, API key, 401 Unauthorized, 403 Forbidden, how to connect an AI coding tool to the Acenda API, how to give ChatGPT the Acenda API docs, how to add the Acenda API to Cursor, how to use AI to build an Acenda integration

How to connect an AI coding tool or agent to the Acenda API