cognit

v1.3.0

Published

14 days ago

The open cognitive ecosystem for AI agents

 ██████╗ ██████╗  ██████╗ ███╗   ██╗██╗████████╗   ███████╗██╗   ██╗███╗   ██╗ ██████╗
██╔════╝██╔═══██╗██╔════╝ ████╗  ██║██║╚══██╔══╝   ██╔════╝╚██╗ ██╔╝████╗  ██║██╔════╝
██║     ██║   ██║██║  ███╗██╔██╗ ██║██║   ██║      ███████╗ ╚████╔╝ ██╔██╗ ██║██║
██║     ██║   ██║██║   ██║██║╚██╗██║██║   ██║      ╚════██║  ╚██╔╝  ██║╚██╗██║██║
╚██████╗╚██████╔╝╚██████╔╝██║ ╚████║██║   ██║      ███████║   ██║   ██║ ╚████║╚██████╗
 ╚═════╝ ╚═════╝  ╚═════╝ ╚═╝  ╚═══╝╚═╝   ╚═╝      ╚══════╝   ╚═╝   ╚═╝  ╚═══╝ ╚═════╝

The CLI for the open cognitive ecosystem for AI agents.

cognit is a fork/extension of the skills ecosystem that adds support for cognitive types: skills, agents, and prompts -- defined via SKILL.md, AGENT.md, and PROMPT.md files.

Supports OpenCode, Claude Code, Codex, Cursor, and 35 more.

Install a Cognitive

npx cognit add vercel-labs/agent-skills

Source Formats

# GitHub shorthand (owner/repo)
npx cognit add vercel-labs/agent-skills

# Full GitHub URL
npx cognit add https://github.com/vercel-labs/agent-skills

# Direct path to a skill in a repo
npx cognit add https://github.com/vercel-labs/agent-skills/tree/main/skills/web-design-guidelines

# GitLab URL
npx cognit add https://gitlab.com/org/repo

# Any git URL
npx cognit add [email protected]:vercel-labs/agent-skills.git

# Local path
npx cognit add ./my-local-skills

Options

| Option | Description | | ------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- | | -g, --global | Install to user directory instead of project | | -a, --agent <agents...> | Target specific agents (e.g., claude-code, codex). See Available Agents | | -s, --skill <skills...> | Install specific cognitives by name (use '*' for all) | | -t, --type <type> | Cognitive type: skill (default), agent, or prompt | | -l, --list | List available cognitives without installing | | -y, --yes | Skip all confirmation prompts | | --all | Install all cognitives to all agents without prompts | | --full-depth | Search all directories recursively for cognitives |

Examples

# List cognitives in a repository
npx cognit add vercel-labs/agent-skills --list

# Install specific cognitives
npx cognit add vercel-labs/agent-skills --skill frontend-design --skill skill-creator

# Install a cognitive with spaces in the name (must be quoted)
npx cognit add owner/repo --skill "Convex Best Practices"

# Install to specific agents
npx cognit add vercel-labs/agent-skills -a claude-code -a opencode

# Non-interactive installation (CI/CD friendly)
npx cognit add vercel-labs/agent-skills --skill frontend-design -g -a claude-code -y

# Install all cognitives from a repo to all agents
npx cognit add vercel-labs/agent-skills --all

# Install all cognitives to specific agents
npx cognit add vercel-labs/agent-skills --skill '*' -a claude-code

# Install specific cognitives to all agents
npx cognit add vercel-labs/agent-skills --agent '*' --skill frontend-design

Installation Scope

| Scope | Flag | Location | Use Case | | ----------- | --------- | ------------------- | --------------------------------------------- | | Project | (default) | ./<agent>/skills/ | Committed with your project, shared with team | | Global | -g | ~/<agent>/skills/ | Available across all projects |

Installation Methods

When installing interactively, you can choose:

| Method | Description | | ------------------------- | ------------------------------------------------------------------------------------------- | | Symlink (Recommended) | Creates symlinks from each agent to a canonical copy. Single source of truth, easy updates. | | Copy | Creates independent copies for each agent. Use when symlinks aren't supported. |

Other Commands

| Command | Description | | ------------------------- | ------------------------------------------------------ | | npx cognit list | List installed cognitives (alias: ls) | | npx cognit find [query] | Search for cognitives interactively or by keyword | | npx cognit remove [names] | Remove installed cognitives from agents | | npx cognit check | Check for available cognitive updates | | npx cognit update | Update all installed cognitives to latest versions | | npx cognit init [name] | Create a new SKILL.md, AGENT.md, or PROMPT.md template |

`cognit list`

List all installed cognitives. Similar to npm ls.

# List all installed cognitives (project and global)
npx cognit list

# List only global cognitives
npx cognit ls -g

# Filter by specific agents
npx cognit ls -a claude-code -a cursor

`cognit find`

Search for cognitives interactively or by keyword.

# Interactive search (fzf-style)
npx cognit find

# Search by keyword
npx cognit find typescript

`cognit check` / `cognit update`

# Check if any installed cognitives have updates
npx cognit check

# Update all cognitives to latest versions
npx cognit update

`cognit init`

# Create SKILL.md in current directory
npx cognit init

# Create a new skill in a subdirectory
npx cognit init my-skill

# Create an AGENT.md or PROMPT.md instead
npx cognit init my-agent -t agent
npx cognit init my-prompt -t prompt

`cognit remove`

Remove installed cognitives from agents.

# Remove interactively (select from installed cognitives)
npx cognit remove

# Remove specific cognitive by name
npx cognit remove web-design-guidelines

# Remove multiple cognitives
npx cognit remove frontend-design web-design-guidelines

# Remove from global scope
npx cognit remove --global web-design-guidelines

# Remove from specific agents only
npx cognit remove --agent claude-code cursor my-skill

# Remove all installed cognitives without confirmation
npx cognit remove --all

# Remove all cognitives from a specific agent
npx cognit remove --skill '*' -a cursor

# Remove a specific cognitive from all agents
npx cognit remove my-skill --agent '*'

# Use 'rm' alias
npx cognit rm my-skill

| Option | Description | | -------------- | ------------------------------------------------ | | -g, --global | Remove from global scope (~/) instead of project | | -a, --agent | Remove from specific agents (use '*' for all) | | -s, --skill | Specify cognitives to remove (use '*' for all) | | -y, --yes | Skip confirmation prompts | | --all | Shorthand for --skill '*' --agent '*' -y |

What are Cognitives?

Cognitives are reusable instruction sets that extend your coding agent's capabilities. cognit supports three cognitive types:

Skills (SKILL.md) -- Task-oriented instructions that teach agents how to perform specific tasks
Agents (AGENT.md) -- Persona and behavior definitions that shape how an agent operates
Prompts (PROMPT.md) -- Reusable prompt templates for common workflows

Each cognitive is defined in its respective markdown file with YAML frontmatter containing a name and description.

Cognitives let agents perform specialized tasks like:

Generating release notes from git history
Creating PRs following your team's conventions
Integrating with external tools (Linear, Notion, etc.)

Discover skills at skills.sh (compatible with cognit)

Supported Agents

Cognitives can be installed to any of these agents:

| Agent | --agent | -------------------------- | --------------------------- | Amp, Kimi Code CLI, Replit | amp, | Antigravity | antigravity | Augment | augment | Claude Code | claude-code | OpenClaw | openclaw | Cline | cline | CodeBuddy | codebuddy | Codex | codex | Command Code | command-code | Continue | continue | Crush | crush | Cursor | cursor | Droid | droid | Gemini CLI | gemini-cli | GitHub Copilot | github-copilot | Goose | goose | Junie | junie | iFlow CLI | iflow-cli | Kilo Code | kilo | Kiro CLI | kiro-cli | Kode | kode | MCPJam | mcpjam | Mistral Vibe | mistral-vibe | Mux | mux | OpenCode | opencode | OpenHands | openhands | Pi | pi | Qoder | qoder | Qwen Code | qwen-code | Roo Code | roo | Trae | trae | Trae CN | trae-cn | Windsurf | windsurf | Zencoder | zencoder | Neovate | neovate | Pochi | pochi | AdaL | adal | Project Path | Global Path | | ---------------------- | ------------------------------- | kimi-cli, replit | .agents/skills/ | ~/.config/agents/skills/ | | .agent/skills/ | ~/.gemini/antigravity/skills/ | | .augment/skills/ | ~/.augment/skills/ | | .claude/skills/ | ~/.claude/skills/ | | skills/ | ~/.moltbot/skills/ | | .cline/skills/ | ~/.cline/skills/ | | .codebuddy/skills/ | ~/.codebuddy/skills/ | | .agents/skills/ | ~/.codex/skills/ | | .commandcode/skills/ | ~/.commandcode/skills/ | | .continue/skills/ | ~/.continue/skills/ | | .crush/skills/ | ~/.config/crush/skills/ | | .cursor/skills/ | ~/.cursor/skills/ | | .factory/skills/ | ~/.factory/skills/ | | .agents/skills/ | ~/.gemini/skills/ | | .agents/skills/ | ~/.copilot/skills/ | | .goose/skills/ | ~/.config/goose/skills/ | | .junie/skills/ | ~/.junie/skills/ | | .iflow/skills/ | ~/.iflow/skills/ | | .kilocode/skills/ | ~/.kilocode/skills/ | | .kiro/skills/ | ~/.kiro/skills/ | | .kode/skills/ | ~/.kode/skills/ | | .mcpjam/skills/ | ~/.mcpjam/skills/ | | .vibe/skills/ | ~/.vibe/skills/ | | .mux/skills/ | ~/.mux/skills/ | | .agents/skills/ | ~/.config/opencode/skills/ | | .openhands/skills/ | ~/.openhands/skills/ | | .pi/skills/ | ~/.pi/agent/skills/ | | .qoder/skills/ | ~/.qoder/skills/ | | .qwen/skills/ | ~/.qwen/skills/ | | .roo/skills/ | ~/.roo/skills/ | | .trae/skills/ | ~/.trae/skills/ | | .trae/skills/ | ~/.trae-cn/skills/ | | .windsurf/skills/ | ~/.codeium/windsurf/skills/ | | .zencoder/skills/ | ~/.zencoder/skills/ | | .neovate/skills/ | ~/.neovate/skills/ | | .pochi/skills/ | ~/.pochi/skills/ | | .adal/skills/ | ~/.adal/skills/ |

[!NOTE] Kiro CLI users: After installing skills, manually add them to your custom agent's resources in .kiro/agents/<agent>.json:
{
  "resources": ["skill://.kiro/skills/**/SKILL.md"]
}

The CLI automatically detects which coding agents you have installed. If none are detected, you'll be prompted to select which agents to install to.

Creating Cognitives

Cognitives are directories containing a SKILL.md, AGENT.md, or PROMPT.md file with YAML frontmatter:

---
name: my-skill
description: What this skill does and when to use it
---

# My Skill

Instructions for the agent to follow when this skill is activated.

## When to Use

Describe the scenarios where this skill should be used.

## Steps

1. First, do this
2. Then, do that

Required Fields

name: Unique identifier (lowercase, hyphens allowed)
description: Brief explanation of what the skill does

Optional Fields

metadata.internal: Set to true to hide the cognitive from normal discovery. Internal cognitives are only visible and installable when INSTALL_INTERNAL_SKILLS=1 is set. Useful for work-in-progress cognitives or those meant only for internal tooling.

---
name: my-internal-skill
description: An internal skill not shown by default
metadata:
  internal: true
---

Cognitive Discovery

The CLI searches for cognitives (SKILL.md, AGENT.md, PROMPT.md) in these locations within a repository:

Root directory (if it contains SKILL.md)
skills/
skills/.curated/
skills/.experimental/
skills/.system/
.agents/skills/
.agent/skills/
.augment/skills/
.claude/skills/
./skills/
.cline/skills/
.codebuddy/skills/
.commandcode/skills/
.continue/skills/
.crush/skills/
.cursor/skills/
.factory/skills/
.goose/skills/
.junie/skills/
.iflow/skills/
.kilocode/skills/
.kiro/skills/
.kode/skills/
.mcpjam/skills/
.vibe/skills/
.mux/skills/
.openhands/skills/
.pi/skills/
.qoder/skills/
.qwen/skills/
.roo/skills/
.trae/skills/
.windsurf/skills/
.zencoder/skills/
.neovate/skills/
.pochi/skills/
.adal/skills/

Plugin Manifest Discovery

If .claude-plugin/marketplace.json or .claude-plugin/plugin.json exists, cognitives declared in those files are also discovered:

// .claude-plugin/marketplace.json
{
  "metadata": { "pluginRoot": "./plugins" },
  "plugins": [
    {
      "name": "my-plugin",
      "source": "my-plugin",
      "skills": ["./skills/review", "./skills/test"]
    }
  ]
}

This enables compatibility with the Claude Code plugin marketplace ecosystem.

If no cognitives are found in standard locations, a recursive search is performed. Use --full-depth to force a full recursive search.

Compatibility

Cognitives are generally compatible across agents since they follow a shared Agent Skills specification. However, some features may be agent-specific:

| Feature | OpenCode | OpenHands | Claude Code | Cline | CodeBuddy | Codex | Command Code | Kiro CLI | Cursor | Antigravity | Roo Code | Github Copilot | Amp | Clawdbot | Neovate | Pi | Qoder | Zencoder | | --------------- | -------- | --------- | ----------- | ----- | --------- | ----- | ------------ | -------- | ------ | ----------- | -------- | -------------- | --- | -------- | ------- | --- | ----- | -------- | | Basic skills | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | | allowed-tools | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | | context: fork | No | No | Yes | No | No | No | No | No | No | No | No | No | No | No | No | No | No | No | | Hooks | No | No | Yes | Yes | No | No | No | No | No | No | No | No | No | No | No | No | No | No |

Troubleshooting

"No cognitives found"

Ensure the repository contains valid SKILL.md, AGENT.md, or PROMPT.md files with both name and description in the frontmatter.

Cognitive not loading in agent

Verify the cognitive was installed to the correct path
Check the agent's documentation for skill/cognitive loading requirements
Ensure the frontmatter is valid YAML

Permission errors

Ensure you have write access to the target directory.

Environment Variables

| Variable | Description | | ------------------------- | ------------------------------------------------------------------------------ | | INSTALL_INTERNAL_SKILLS | Set to 1 or true to show and install cognitives marked as internal: true | | DISABLE_TELEMETRY | Set to disable anonymous usage telemetry | | DO_NOT_TRACK | Alternative way to disable telemetry |

# Install internal cognitives
INSTALL_INTERNAL_SKILLS=1 npx cognit add vercel-labs/agent-skills --list

Telemetry

This CLI collects anonymous usage data to help improve the tool. No personal information is collected.

Telemetry is automatically disabled in CI environments.

License

MIT