agentleFS
Sign inStart free
Cognitic-LabsCognitic-Labs/geoskills/AGENTS.md

geoskills

AGENTS.md30 starsChanged 43 days agoApache-2.0View on GitHub

geoskills is a multi-skill monorepo for Generative Engine Optimization (GEO). Each skill follows the open AgentSkills standard and is compatible with Claude Code, OpenCode, OpenClaw, Codex CLI, Cursor, GitHub Copilot, and other AgentSkills-compatible agents. Every skill under skills/ MUST follow this structure:

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,700 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository9 other files it gives its agents.

The file

We split this AGENTS.md into the rules it gives an agent

10 rules. About 0 are common, 6 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Every skill under skills/ MUST follow this structure:

    line 81
  2. Distinctive:

    Formatting: Use tables for structured data, code blocks for examples

    line 116
  3. Distinctive:

    Scoring: Always use 0-100 scale with sub-dimension breakdowns

    line 117
  4. Very specific:

    Commit messages: Conventional Commits format (feat:, fix:, docs:, refactor:)

    line 124
  5. Distinctive:

    Always update root README "Available Skills" table when adding a new skill

    line 126
  6. Very specific:

    Add SKILL.md with proper frontmatter (name, description, version)

    line 131
  7. Distinctive:

    Add README.md with skill documentation

    line 132
  8. Very specific:

    Add references/, scripts/, evals/ as needed

    line 133
  9. Distinctive:

    Update root README.md Available Skills table

    line 135
  10. Distinctive:

    Update this file's directory structure diagram

    line 136

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

geoskills — AI Agent Guide

Repository Overview

geoskills is a multi-skill monorepo for Generative Engine Optimization (GEO). Each skill follows the open AgentSkills standard and is compatible with Claude Code, OpenCode, OpenClaw, Codex CLI, Cursor, GitHub Copilot, and other AgentSkills-compatible agents.

Directory Structure

geoskills/
├── .gitignore
├── AGENTS.md                       ← You are here
├── README.md                       # Monorepo-level docs
├── LICENSE
├── package.json
├── skills/
│   ├── geo-audit/                  # Full GEO audit skill
│   │   ├── SKILL.md                # Skill definition (frontmatter + prompt)
│   │   ├── README.md               # Skill documentation
│   │   ├── references/             # Scoring rubrics, data & subagent definitions
│   │   │   ├── scoring-guide.md
│   │   │   └── agents/             # Subagent instruction files
│   │   │       ├── geo-technical.md
│   │   │       ├── geo-citability.md
│   │   │       ├── geo-schema.md
│   │   │       └── geo-brand.md
│   │   └── evals/                  # Evaluation test cases
│   │       └── evals.json
│   ├── geo-fix-llmstxt/            # llms.txt generator skill
│   │   ├── SKILL.md
│   │   ├── README.md
│   │   └── references/
│   │       └── llmstxt-spec.md     # llms.txt specification reference
│   ├── geo-fix-schema/             # JSON-LD schema generator skill
│   │   ├── SKILL.md
│   │   ├── README.md
│   │   └── references/
│   │       └── schema-templates.md # JSON-LD template patterns
│   ├── geo-fix-content/            # Content citability rewriter skill
│   │   ├── SKILL.md
│   │   ├── README.md
│   │   └── references/
│   │       ├── hedge-words.md      # Hedge language dictionary
│   │       └── quotable-content-examples.md
│   ├── geo-compare/                # Competitor comparison skill
│   │   ├── SKILL.md
│   │   └── README.md
│   ├── geo-monitor/                # Score tracking over time skill
│   │   ├── SKILL.md
│   │   └── README.md
│   ├── seo-analytics/              # SEO analytics from GSC, Bing Webmaster Tools, and GA4
│   │   ├── SKILL.md
│   │   ├── README.md
│   │   ├── references/
│   │   │   ├── config.md
│   │   │   ├── event-model.md
│   │   │   ├── localization.md
│   │   │   └── report-outline.md
│   │   ├── scripts/
│   │   │   ├── fetch-gsc.mjs
│   │   │   ├── fetch-bing-webmaster.mjs
│   │   │   ├── fetch-ga4.mjs
│   │   │   └── lib/
│   │   │       └── cli-utils.mjs
│   │   └── evals/
│   │       └── evals.json
│   └── bing-ai-analytics/          # Bing AI Performance export analytics
│       ├── SKILL.md
│       ├── README.md
│       ├── references/
│       │   ├── export-schema.md
│       │   ├── recommendation-handling.md
│       │   └── report-outline.md
│       └── evals/
│           └── evals.json
└── raw/                            # Research data (not a skill)

Skill Internal Structure

Every skill under skills/ MUST follow this structure:

skills/<skill-name>/
├── SKILL.md          # Required — skill definition with frontmatter
├── README.md         # Required — human-readable documentation
├── references/       # Optional — scoring guides, rubrics, subagent instructions
│   └── agents/       # Optional — subagent instruction files
├── scripts/          # Optional — executable helper scripts
└── evals/            # Optional — evaluation test cases
    └── evals.json

SKILL.md Conventions

  • Frontmatter is YAML between --- fences at the top
  • Required fields: name, description
  • Recommended fields: version (semver, e.g., 1.0.0)
  • name MUST match the directory name (e.g., skills/geo-audit/ → name: geo-audit)
  • description should include trigger phrases (20-40 words), use "Use when..." pattern
  • Body contains the full system prompt for the skill
  • Body MUST use tool-agnostic natural language (no platform-specific tool names like WebFetch, Bash, Agent)

Subagent Instruction Files (references/agents/*.md)

  • Each file defines one subagent's instructions
  • Frontmatter fields: name, description
  • name format: geo-<domain> (e.g., geo-technical, geo-citability)
  • Body contains the subagent's system prompt with scoring rubrics
  • Instructions must be tool-agnostic (no platform-specific tool names)

Writing Style Guide

  • Tone: Professional, technical, concise
  • Audience: Developers and SEO professionals
  • Formatting: Use tables for structured data, code blocks for examples
  • Scoring: Always use 0-100 scale with sub-dimension breakdowns
  • References: Cite research sources when making claims
  • No emojis unless explicitly part of a UI element

Git Workflow

  • Branch naming: feat/<skill-name>-<description>, fix/<skill-name>-<description>
  • Commit messages: Conventional Commits format (feat:, fix:, docs:, refactor:)
  • One skill per PR when adding new skills
  • Always update root README "Available Skills" table when adding a new skill

Adding a New Skill

  1. Create skills/<skill-name>/ directory
  2. Add SKILL.md with proper frontmatter (name, description, version)
  3. Add README.md with skill documentation
  4. Add references/, scripts/, evals/ as needed
  5. Subagent instructions go in references/agents/ (not a top-level agents/ dir)
  6. Update root README.md Available Skills table
  7. Update this file's directory structure diagram

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free