We're excited to announce that Transloadit has joined the growing movement to make the web more AI-friendly by implementing support for /llms.txt and /llms-full.txt files. Our extensive documentation and API capabilities are now more accessible to large language models (LLMs), enabling developers to build more powerful AI-integrated applications with Transloadit.

What is llms.txt?

Proposed in September 2024 by Jeremy Howard, co-founder of Answer.AI, the /llms.txt standard addresses a critical challenge facing AI systems: hallucinations and context window limitations. The file provides structured, concise information that is specifically formatted for LLMs.

Similar to how robots.txt helps search engines to navigate websites, llms.txt guides LLMs. It eliminates the need for AI to parse complex HTML, JavaScript, and other web elements, which consumes valuable context window space. These files offer a high signal-to-noise ratio, helping LLMs produce better results without relying on compressed – and potentially outdated – information.

The standard consists of two main components:

  • /llms.txt: a concise overview with structured links to key resources.
  • /llms-full.txt: a comprehensive version containing detailed documentation in markdown format.

Why does this matters to developers?

Transloadit's implementation of llms.txt offers significant benefits for developers working with AI tools:

  1. Improved AI understanding: AI assistants can now better understand Transloadit's capabilities. This makes it easier to generate accurate code examples and explanations when building integrations.

  2. Efficient context usage: the streamlined format allows you to include Transloadit's documentation in your AI prompts, without wasting valuable context tokens.

  3. More accurate responses: AI tools can access up-to-date information about our Robots, parameters, and workflows, which reduces errors in generated code.

  4. Enhanced developer experience: when using AI coding assistants like GitHub Copilot or Cursor, you'll receive more precise guidance on implementing Transloadit's file processing capabilities.

What is included in our implementation?

Our /llms.txt file provides a concise overview of Transloadit's core functionality, including:

  • a clear explanation of our file uploading and processing capabilities;
  • example Assembly Instructions, demonstrating our JSON-based workflow configuration;
  • links to key documentation resources;
  • a comprehensive list of available Robots.

The more detailed /llms-full.txt expands on this with:

  • detailed Robot parameter schemas;
  • code examples for various integration scenarios;
  • comprehensive documentation on all aspects of the Transloadit service.

How to use llms.txt with AI tools

When working with AI assistants like ChatGPT, Claude, or GitHub Copilot, you can now include our llms.txt files to provide better context. You can directly reference URLs by pointing the AI to https://transloadit.com/llms.txt or https://transloadit.com/llms-full.txt. In Cursor, you can also mention the URL with @https://transloadit.com/llms-full.txt. In some cases, AI agents may even automatically detect the existence of llms.txt on a domain, and include it when needed. Content can then be directly included by copying it into your prompts for local AI processing. When you're done, simply download the files and upload them to AI tools that support file attachments.

Join the AI-friendly web movement

We at Transloadit recognize the importance of making the web more accessible to AI systems, and the llms.txt standard represents an important step toward making that happen. As more services implement this standard, developers will be able to benefit from improved AI assistance across their entire technology stack.

We encourage other services to join this movement and implement their own llms.txt files. For more information about the standard, visit llmstxt.org.

Resources

We're excited to see how developers will use this new capability to build even more powerful applications with Transloadit. As always, we welcome your feedback and questions in our Community Forum.