Skip to main content

​
General Questions

llms.txt is a standard format that helps Large Language Models (LLMs) quickly understand and navigate documentation. Instead of crawling through entire documentation sites, AI tools can read a single, well-structured file that contains all the essential information.For Nigerian fintech APIs, this means:
  • Faster integration with AI coding assistants
  • More accurate code generation
  • Better understanding of API capabilities and limitations
  • Reduced hallucinations from AI tools
This project was born from building the Mono MCP server. During development with coding agents, LLMs struggled to extract relevant information from large documentation sites. Creating structured llms.txt files solved this problem dramatically.The goal is to help Nigerian fintech companies adopt this standard natively on their documentation sites, making it easier for developers to build with AI assistance.
Currently, the collection includes:
  • Mono - Open Banking (financial data access, payments, identity verification)
  • OPay - Payments (gateway with cards, bank transfers, USSD, wallet, POS)
  • Paystack - Payments (processing, transfers, subscriptions, terminals)
Note: Flutterwave already has native llms.txt support through Mintlify.
No, this is an independent, open-source project. The llms.txt files are created by parsing and structuring publicly available documentation.The goal is for these companies to eventually adopt these files natively on their own documentation sites. If you work for one of these companies, we’d love to collaborate!

​
Using llms.txt Files

The usage depends on your AI tool:With Claude/ChatGPT:
  1. Copy the content from the relevant llms.txt file
  2. Paste it into your conversation before asking integration questions
  3. Reference specific sections when asking about API endpoints
With coding agents:
  • Point your agent to the raw GitHub URL of the llms.txt file
  • Configure it as a context source before starting your task
With MCP servers:
  • Use tools like the Mono MCP server that natively understand these files
  • Configure the file path in your MCP server settings
Yes! Clone the GitHub repository and access all files locally:
git clone https://github.com/sin4ch/nigerian-fintech-llms-txt.git
cd nigerian-fintech-llms-txt
Then point your AI tools to the local file paths.
These files work best with:
  • AI coding assistants that support context files (Cursor, GitHub Copilot, etc.)
  • Chat interfaces that accept long-form context (Claude, ChatGPT)
  • MCP servers and other AI agent frameworks
  • Tools that specifically support the llms.txt standard
The files are formatted according to the llms.txt specification, so any tool supporting this standard should work.
We strive to keep these files current, but they may not always reflect the very latest changes. This is another reason why native adoption by fintech companies is important.If you notice outdated information:
  1. Check the official documentation
  2. Open an issue on GitHub with details
  3. Or submit a pull request with the updates

​
Contributing

To add a new provider:
  1. Create a directory with the provider name in the repository
  2. Create an llms.txt file with two main sections:
    • Documentation overview (guides, concepts, SDKs)
    • API reference (endpoints, methods, parameters)
    • Separate sections with ---
  3. Follow the format used in existing files (Mono, OPay, Paystack)
  4. Update the README to include the new provider in the table
  5. Submit a pull request with a clear description
See the Contributing Guide for detailed instructions.
A good llms.txt file should include:Documentation Overview:
  • Getting started guides
  • Core concepts and terminology
  • Authentication methods
  • SDK information
  • Common use cases
API Reference:
  • All API endpoints with descriptions
  • Request/response formats
  • Required and optional parameters
  • Error codes and handling
  • Rate limits and constraints
The goal is to give an LLM everything it needs to help a developer integrate the API.
No permission is needed to contribute, as these files are based on publicly available documentation. The project is open source under the MIT License.However, we encourage:
  • Respecting trademark and branding guidelines
  • Accurate representation of the APIs
  • Proper attribution to the original documentation
The ultimate goal is for companies to adopt these files natively.
This project is licensed under the MIT License. This means:
  • You can use these files freely
  • You can modify and distribute them
  • You can use them in commercial projects
  • No warranty is provided
See the LICENSE file for full details.
Absolutely! That’s the goal. If you represent a fintech company:
  1. You can adopt your llms.txt file directly on your documentation site
  2. You can modify it to better suit your needs
  3. You can use tools like Mintlify to auto-generate and maintain it
We’d love to hear from you if you decide to implement it natively!
Have a question not answered here? Ask on GitHub Discussions or open an issue.

Build docs developers (and LLMs) love

Get started for freeTalk to us