How real projects summarise themselves for language models.
hnikoloskillms.txthnikoloski/imlazy/docs/llms.txt
verify gate - Standard tier: 1-3 file features/bugfixes with clear requirements — light planning, TDD, code review - Heavy tier: multi-component features, ambiguous scope, architectural design — brainstorming, full planning with subagents
infinitegameplayerllms.txtinfinitegameplayer/sovereign-ecosystem/llms.txt
Manuscript Anti-AI Edit Pass: https://infinitegameos.io/skills/anti-ai-edit-pass - Systematic Debugging: https://infinitegameos.io/skills/systematic-debugging - PR Code Review: https://infinitegameos.io/skills/pr-code-review - Autonomous Improvement Session: https://infinitegameos.io/skills/autonomous-improvement-session - Batch Archival: https://infinitegameos.io/skills/batch-archival
jasonmichaelbell78-creatorllms.txtjasonmichaelbell78-creator/sonash-v0/llms.txt
single-session AI optimization audit on the codebase - [/audit-code](.claude/skills/audit-code/SKILL.md): Run a single-session code review audit on the codebase - [/audit-comprehensive](.claude/skills/audit-comprehensive/SKILL.md): Run all 9 domain audits in staged waves
jeonnoin-altllms.txtjeonnoin-alt/Eureka/llms.txt
scientific discipline on AI coding agents the way Superpowers enforces test-driven development and code review. It ships as a set of composable "skills" (auto-triggering workflow prompts) plus fresh
kina2711llms.txtkina2711/data-department-agent-skills/skills/llms.txt
with a fixed glossary, style guide, fidelity review and translation memory. Produces: localised content; translation memory; transcreated copy; translated code documentation; translated document; translated book manuscript. Task groups: build-deliver
koryphllms.txtkoryph/koryph/docs/llms.txt
toolkit (`inject`, `status --frontier`, the overrides sidecar). - [AI runtimes: support status](https://github.com/koryph/koryph/blob/main/docs/user-guide/runtimes.md): Claude Code is the production runtime; all other runtimes (codex, cursor, gemini, grok, copilot, opencode
LARIkozllms.txtLARIkoz/ai-model-benchmarks/llms-full.txt
Best Model | Backup | Free | |------|-----------|--------|------| | Write/review code | Claude Sonnet 4.6 | MiniMax M2.5 | MiniMax M2.5 FREE | | Code review (post-implement) | Codex CLI | Qwen CLI | None | | Spec/architecture review | GPT-5.4 | Grok 4.20 | None
LARIkozllms.txtLARIkoz/kurdyuk-lite/llms.txt
Code, Codex, Gemini, and any agent that reads a project instructions file). It replaces free-form chat with a strict six-phase pipeline — Brief → Charter → Spec → Plan → Implement → Final Review
lm203688llms.txtlm203688/aishield/docs/llms-full.txt
deduplicated, zero false positives on a benign corpus). **Deployment:** fully local / offline. No source code is uploaded to any cloud. Detection runs on the developer's machine
lordbasilaiassistant-sudollms.txtlordbasilaiassistant-sudo/ai-company-bible/llms-full.txt
ledger.** A shape check proves well-formed, never true. If no line of code would reject the well-formed lie, the word "verified" is banned from the copy. (Two-gates
lpalboullms.txtlpalbou/AgoraHub/llms.txt
task assignments, dependency/reply waits, VFS byte digests and finding integration, delivery preparation, current typed reviews, routing and briefings. ## Core Docs - [README.md](README.md): project overview, quick start, and how agents connect
MarkDonishllms.txtMarkDonish/round-table-workspace/docs/llms.txt
Round Table Workspace Round Table Workspace is a local-first review workflow for AI coding agents before trusting generated work. Primary repo: https://github.com/MarkDonish/round-table-workspace Pages overview: https://markdonish.github.io/round-table-workspace/
MaururuTakumillms.txtMaururuTakumi/claude-agmsg-delegate/llms.txt
review and opt-in Sonnet Git-workspace implementation through agmsg, restricted to a verified paid Claude.ai subscription. Repository: https://github.com/MaururuTakumi/claude-agmsg-delegate License: MIT Runtime dependencies: Python 3.9+, Codex, Claude Code
mingyo186llms.txtmingyo186/partreel/llms.txt
feedback via GitHub: https://github.com/mingyo186/partreel New parts are auto-reviewed by the same quality gates in CI. ## License Code MIT. Component assets CC-BY-4.0 (attribution: "PartReel
MSKazemillms.txtMSKazemi/idkmesh/docs/llms.txt
guides. - [AI agent verification](https://mskazemi.com/idkmesh/topics/ai-agent-verification.html) - [Multi-agent orchestration](https://mskazemi.com/idkmesh/topics/multi-agent-orchestration.html) - [AI code review](https://mskazemi.com/idkmesh/topics/ai-code-review.html) - [LLM-as-a-judge reliability](https://mskazemi.com/idkmesh/topics/llm-judge-reliability.html) - [Verifier panels
MuscleOtterllms.txtMuscleOtter/rowan-fitness-skill/llms.txt
skill support and the Project fallback. A loaded skill does not itself establish independent reviewers, durable saves or device access. ## Start and identify - [Project overview](https://raw.githubusercontent.com/MuscleOtter/rowan-fitness-skill/main/README.md): Audience, goals
NAME0x0llms.txtNAME0x0/OpenEchelon/llms.txt
employee can be run by a local model, a frontier API, or a CLI coding agent depending on the task. OpenEchelon is licensed under Apache-2.0 and is currently
NissanOhanallms.txtNissanOhana/orangu/site/llms-full.txt
observability for Claude Code, Cowork, and Desktop sessions, with optional Claude Code skills that turn the evidence into reviewable changes. [sample report](https://nissanohana.github.io/orangu/sample.html) · [repo sample](https://nissanohana.github.io/orangu/sample-repo.html)
NissanOhanallms.txtNissanOhana/orangu/site/llms.txt
your config declares vs what your sessions actually used). Optional Claude Code skills turn > that evidence into reviewable proposals and receipts. The loop is observe -> see -> improve. Concretely
Nizokallms.txtNizoka/pdfnative-mcp/llms.txt
validate a Tagged PDF for PDF/UA (ISO 14289-1) structural accessibility, - embed a QR / Code 128 / EAN-13 / Data Matrix / PDF417 barcode, - render text in any of 27 Unicode scripts
A markdown file at a website's root that gives language models a short guide to the site and links to the pages worth reading.
The same idea with the content included, so an agent can read the documentation in one request.
Agents and tools that fetch documentation on someone's behalf. It's a proposal, not a standard, and support varies.
Start with a one-line summary, then sections of links with a sentence each. The examples here show what real projects do.
