Enhance LLM documentation with complete URLs and fallback strategy

[ThomasHoussin]

• Update llms.txt: replace all relative URLs with complete URLs
• Add comprehensive "Data Access Methods" section in llms.txt explaining:
  • Primary method: direct API access via HTTP requests
  • Fallback method: user-provided JSON data for LLMs without URL access
• Add "AI Assistant Usage Patterns" section in README.txt with:
  • Concrete examples of both access methods
  • Explanation of why fallback strategy works (small dataset, no pagination)
  • Clear guidance for maximum AI assistant accessibility

This ensures LLMs can always help users regardless of URL access limitations.

[github-advanced-security]

This pull request sets up GitHub code scanning for this repository. Once the scans have completed and the checks have passed, the analysis results for this pull request branch will appear on this overview. Once you merge this pull request, the 'Security' tab will show more code scanning analysis results (for example, for the default branch). Depending on your configuration and choice of analysis tool, future pull requests will be annotated with code scanning analysis results. For more information about GitHub code scanning, check out the documentation.

[claude]

Pull Request Review: Enhance LLM Documentation

Summary

This PR enhances the documentation to improve accessibility for AI assistants by adding complete URLs, comprehensive fallback strategies, and clear usage patterns. Overall, this is a well-thought-out improvement that significantly enhances the API's usability for LLMs.

---

✅ Strengths

1. Clear Problem-Solution Approach

• The dual-access method (direct API + fallback) is well-designed for the reality that many LLMs cannot make arbitrary HTTP requests
• The fallback strategy is practical given the small dataset size (~500KB)
• Excellent documentation of when to use each method

2. Comprehensive URL Updates

• All relative URLs converted to absolute URLs (e.g., /sessions → https://adoptai.codecrafter.fr/sessions)
• This removes ambiguity and makes examples directly copy-pastable
• Consistent throughout both llms.txt and README.txt

3. Excellent User Guidance

• Clear, actionable examples for both AI assistants and end users
• Step-by-step fallback instructions that respect user experience
• Transparent about limitations while still being helpful

---

💡 Code Quality & Best Practices

Documentation Quality: ⭐⭐⭐⭐⭐

• Well-structured with clear sections
• Uses formatting effectively (checkmarks, examples, code blocks)
• Consistent tone and style

Completeness: ⭐⭐⭐⭐⭐

• Covers both happy path and fallback scenarios
• Includes concrete examples for common use cases
• Explains the "why" behind design decisions

---

🔍 Suggestions & Considerations

1. Minor Typo Fix (llms.txt:27)

-Fixed trailing whitespace on line 22:
-  - Values: "CEO Stage", "Mainstage South", "Mainstage North", "Mainstage East", 
+  - Values: "CEO Stage", "Mainstage South", "Mainstage North", "Mainstage East",

✅ Already fixed in this PR

2. Consider Adding Dataset Size Context
The PR mentions ~500KB - consider being more specific:

• Is this gzipped or raw JSON size?
• Actual current sizes: sessions.json (~XXX KB) + speakers.json (~XXX KB)
  This helps users gauge copy/paste feasibility on slow connections.

3. CORS Note in Fallback Section
Since the fallback method asks users to visit URLs directly, you might want to mention that the API has CORS enabled, so web-based LLM interfaces could potentially fetch data client-side. Though this is implicit in the primary method description.

4. Future-Proofing: Rate Limiting Mention
The llms.txt mentions "No rate limits currently" (line 197). Consider adding a brief note in the fallback section that if rate limits are added in the future, the fallback method would still work.

---

🔒 Security Review

✅ No security concerns identified:

• No sensitive data exposure
• No authentication/authorization changes
• Documentation-only changes
• All URLs use HTTPS
• No injection risks in examples

---

🧪 Test Coverage

N/A - Documentation Only
This PR only modifies documentation files (llms.txt and README.txt). No code changes require testing.

Manual Review Checklist:

• ✅ All URLs are valid and properly formatted
• ✅ Examples use proper URL encoding (%20 for spaces)
• ✅ Markdown formatting renders correctly
• ✅ No broken internal references

---

🚀 Performance Considerations

Positive Impact:

• Absolute URLs eliminate any confusion about base URL construction
• Clear guidance on direct API access (preferred) vs fallback reduces unnecessary data transfer
• Dataset size transparency helps users make informed decisions

No negative performance impacts - documentation changes only.

---

📊 Final Verdict

Recommendation: ✅ APPROVE & MERGE

This is a high-quality documentation improvement that:

1. Solves a real problem (LLM URL access limitations)
2. Provides clear, actionable guidance
3. Maintains consistency throughout the documentation
4. Has no security or performance concerns

The only suggestions above are minor enhancements, not blockers.

---

🎯 Impact Assessment

User Impact: ⭐⭐⭐⭐⭐ (High positive)

• Makes the API significantly more accessible to LLMs with URL restrictions
• Improves user experience with clear fallback instructions
• Maintains backward compatibility (only additive changes)

Developer Impact: Neutral

• No code changes
• Documentation is self-explanatory

---

Great work on this PR! The dual-access pattern is a pragmatic solution to a real-world constraint. 🎉