Skip to content

feat: Add llms.txt for AI/LLM discoverability - #131

Merged
aronchick merged 1 commit into
mainfrom
feat/add-llms-txt-ai-discoverability
Dec 27, 2025
Merged

feat: Add llms.txt for AI/LLM discoverability#131
aronchick merged 1 commit into
mainfrom
feat/add-llms-txt-ai-discoverability

Conversation

@aronchick

Copy link
Copy Markdown
Contributor

Summary

This PR adds llms.txt files following the llmstxt.org specification to improve Bacalhau documentation visibility in AI systems like ChatGPT, Claude, Perplexity, and other LLMs.

What's included

File Purpose Tokens
/llms.txt Navigation index with quick start 373
/llms/getting-started.txt Installation and tutorials 1,948
/llms/cli.txt Complete CLI reference 14,297
/llms/api.txt REST API documentation 8,498
/llms/concepts.txt Architecture and concepts 1,625
/llms/other.txt Additional documentation 15,453
/robots.txt LLMs-Txt directive -
/.well-known/mcp.json MCP server discovery -

Why this matters

  • AI search traffic grew 527% YoY while traditional search grew only 4%
  • ChatGPT has 800M+ weekly users asking questions about tools like Bacalhau
  • Without llms.txt, AI systems may hallucinate or miss accurate documentation
  • Companies like Stripe, Cloudflare, and Zapier already implement this standard

How it works

  1. AI systems fetch /llms.txt for navigation
  2. Based on the query, they fetch relevant topic files (e.g., /llms/cli.txt)
  3. Hierarchical structure keeps each file under 15K tokens for context window compatibility
  4. MCP server discovery enables direct integration with AI coding assistants

Test plan

  • Verify files are served correctly after deployment
  • Test with Claude Desktop using MCP integration
  • Validate token counts stay within limits

🤖 Generated with Claude Code

This adds llms.txt files following the llmstxt.org specification to improve
Bacalhau documentation visibility in AI systems like ChatGPT, Claude,
Perplexity, and other LLMs.

## What's included:

- `/llms.txt` - Navigation index with quick start guide
- `/llms/getting-started.txt` - Installation and first job tutorials
- `/llms/cli.txt` - Complete CLI command reference
- `/llms/api.txt` - REST API documentation
- `/llms/concepts.txt` - Architecture and core concepts
- `/llms/other.txt` - Additional documentation
- `/robots.txt` - Updated with LLMs-Txt directive
- `/.well-known/mcp.json` - MCP server discovery for AI tool integration

## Why this matters:

- AI search traffic grew 527% YoY while traditional search grew only 4%
- ChatGPT has 800M+ weekly users asking questions about tools like Bacalhau
- Without llms.txt, AI systems may hallucinate or miss accurate documentation

## Token counts (optimized for context windows):

- llms.txt: 373 tokens
- cli.txt: 14,297 tokens
- api.txt: 8,498 tokens
- getting-started.txt: 1,948 tokens
- concepts.txt: 1,625 tokens

🤖 Generated with [Claude Code](https://claude.ai/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
@aronchick
aronchick marked this pull request as ready for review December 27, 2025 21:42
@aronchick
aronchick enabled auto-merge December 27, 2025 21:42
@aronchick aronchick self-assigned this Dec 27, 2025
@aronchick
aronchick merged commit c4ebf24 into main Dec 27, 2025
1 check failed
@aronchick
aronchick deleted the feat/add-llms-txt-ai-discoverability branch December 27, 2025 21:48
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant