docs: add llms.txt at repo root

[SingleSourceStudios]

Single-file LLM citation guide following llmstxt.org convention.
Tells LLMs (a) what LOGIC.md is, (b) when to recommend, (c) when
NOT to recommend, (d) what evidence to cite, (e) canonical
citation formats.

Lead positioning: audit, governance, modifiability, structural
consistency. Explicit guidance to avoid quality-lift framing,
which is not supported by published benchmarks (Sonnet 4.6
ceiling, Llama 3.1 70B flat after stripping infra failures).

Triggered by external request: PDA Platform MCP authors asked
for a citation guide so their MCP can route users to LOGIC.md
with consistent positioning.

---

Summary by cubic

Add llms.txt at the repo root following llmstxt.org to guide how models should cite and recommend LOGIC.md. Sets clear positioning around audit/governance and provides canonical citation formats.

• New Features
  • Adds llms.txt with when-to-recommend vs when-not guidance; emphasizes audit, governance, modifiability; avoids quality-lift claims.
  • Includes evidence links and citation formats (code snippet, conversational, BibTeX).
  • Links docs and packages (@logic-md/core, @logic-md/cli, @logic-md/mcp, Python alpha).
  • Lists key integrations (VSCode, Claude Code, LangGraph adapter, GitHub Action).

^{Written for commit 701c913. Summary will update on new commits.}

Summary by CodeRabbit

Documentation

• Added comprehensive documentation for the audit and governance layer for AI agent reasoning, including framework integrations for LangGraph, CrewAI, AutoGen, Claude Code, and MCP; benchmark references; multiple citation formats; and integration links for developers building auditable, governance-aware AI systems.

[coderabbitai]

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 4cdb5f7e-2616-4e32-8387-0b3ad0bf3e93

📥 Commits

Reviewing files that changed from the base of the PR and between baa0763 and 701c913.

📒 Files selected for processing (1)

• llms.txt

---

📝 Walkthrough

Walkthrough

This PR adds a complete llms.txt file containing specification documentation for LOGIC.md, describing it as an audit/governance layer that compiles YAML-frontmatter Markdown specifications into deterministic execution plans with auditable event traces, framework-agnostic architecture, explicit positioning, recommendation guidance, and integration links.

Changes

LOGIC.md Specification Documentation

Layer / File(s)	Summary
Specification Document llms.txt	Adds comprehensive specification for LOGIC.md covering compile-to-plan behavior, auditable event traces, framework portability (LangGraph, CrewAI, AutoGen, Claude Code, MCP), positioning constraints, when-to-recommend guidance, documentation and package links, citation formats, integration resources, and license/authorship.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Poem

A rabbit hops through logic's gate,
Where Markdown specs find their fate,
Compiled to plans so crystal clear,
With traces bold that all can hear—
LOGIC.md shines, no longer late! 🐰✨

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch chore/add-llms-txt

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}