Reskill Usage logo

Reskill Usage

Organization
kanyun-inc
reskill-usage

Teaches AI agents how to use reskill — a Git-based package manager for AI agent skills. Covers CLI commands, install formats, configuration, publishing, and common workflows.

Overview

Publisherkanyun-inc
Repositoryreskill
Skill namereskill-usage
Stars
60
Forks
2
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by kanyun-inc on GitHub. Read the source before you install it.

Installation

Install the Reskill Usage AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/kanyun-inc/reskill/tree/main/skills/reskill-usage
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/kanyun-inc/reskill.git /tmp/reskill
mkdir -p .claude/skills
cp -r /tmp/reskill/skills/reskill-usage .claude/skills/reskill-usage
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Reskill Usage in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Reskill Usage on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Reskill Usage is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

reskill Usage Guide

reskill is a Git-based package manager for AI agent skills. It provides declarative configuration (skills.json + skills.lock), flexible versioning, and multi-agent support for installing, managing, and sharing skills across projects and teams.

Requirements: Node.js >= 18.0.0

CLI usage: If reskill is installed globally, use it directly. Otherwise use npx reskill@latest:

bash
npm install -g reskill        # Global install
npx reskill@latest <command>  # Or use npx directly (no install needed)

When to Use This Skill

Use this skill when the user:

  • Wants to install, update, or manage AI agent skills
  • Mentions skills.json, skills.lock, or reskill-related concepts
  • Wants to publish a skill to a registry
  • Asks about supported install formats (GitHub, GitLab, HTTP, OSS, registry, etc.)
  • Encounters reskill-related errors or needs troubleshooting
  • Wants to set up a project for skill management
  • Asks about multi-agent skill installation (Cursor, Claude Code, Codex, etc.)

AI Agent Execution Rules

AI agents cannot respond to interactive prompts mid-command. Always add -y to commands that support confirmation prompts (install, uninstall, publish) to prevent the command from hanging.

bash
# Correct — will not hang
reskill install github:user/skill -y
reskill uninstall skill-name -y
reskill publish -y

# Wrong — will hang waiting for confirmation
reskill install github:user/skill
reskill uninstall skill-name

Quick Start

bash
# Initialize a new project
npx reskill@latest init

# Install a skill
npx reskill@latest install github:anthropics/skills/skills/frontend-design@latest

# List installed skills
npx reskill@latest list

Commands

CommandAliasDescription
init-Initialize skills.json
find <query>searchSearch for skills in the registry
install [skills...]iInstall one or more skills
listlsList installed skills
info <skill>-Show skill details
update [skill]upUpdate skills
outdated-Check for outdated skills
uninstall <skill>un, rm, removeRemove a skill
group-Manage skill groups ¹
publish [path]pubPublish a skill to the registry ¹
login-Authenticate with the registry ¹
logout-Remove stored authentication ¹
whoami-Display current logged in user ¹
doctor-Diagnose environment and check for issues
completion [action]-Setup or remove shell tab completion

¹ Registry commands (group, publish, login, logout, whoami) require a private registry deployment. Not available for public use yet.

Run reskill <command> --help for complete options and detailed usage.

Common Options

OptionCommandsDescription
--no-saveinstallInstall without saving to skills.json (for personal skills)
-g, --globalinstall, uninstall, listInstall/manage skills globally (user directory)
-a, --agent <agents...>installSpecify target agents (e.g., cursor, claude-code)
-a, --agent <agent>listList skills installed to a specific agent
--mode <mode>installInstallation mode: symlink (default) or copy
--allinstallInstall to all agents
-y, --yesinstall, uninstall, publishSkip confirmation prompts
-f, --forceinstallForce reinstall even if already installed
-s, --skill <names...>installSelect specific skill(s) by name from a multi-skill repo
--listinstallList available skills in the repository without installing
--skip-manifestinstallSkip all skills.json and skills.lock writes (for platform integration)
-t, --token <token>install, find, group, publish, loginAuth token for registry API requests (for CI/CD)
-r, --registry <url>install, find, group, publish, login, logout, whoamiRegistry URL override for registry-enabled commands
--tag <tag>publishGit tag to publish
--access <level>publishAccess level: public (default) or restricted
-n, --dry-runpublishValidate without publishing
-g, --group <path>publishPublish skill into a group (e.g., kanyun/frontend)
-j, --jsonlist, info, outdated, doctor, group, findOutput as JSON
-l, --limit <n>findMaximum number of search results
--skip-networkdoctorSkip network connectivity checks

Source Formats

reskill supports installing skills from multiple sources:

bash
# GitHub shorthand
npx reskill@latest install github:user/skill@v1.0.0

# GitLab shorthand
npx reskill@latest install gitlab:group/skill@latest

# Full Git URL (HTTPS)
npx reskill@latest install https://github.com/user/skill.git

# Full Git URL (SSH)
npx reskill@latest install git@github.com:user/skill.git

# GitHub/GitLab web URL (with branch and subpath)
npx reskill@latest install https://github.com/vercel-labs/agent-skills/tree/main/skills/web-design-guidelines

# Custom registry (self-hosted GitLab, etc.)
npx reskill@latest install gitlab.company.com:team/skill@v1.0.0

# HTTP/OSS archives
npx reskill@latest install https://example.com/skills/my-skill-v1.0.0.tar.gz
npx reskill@latest install oss://bucket/path/skill.tar.gz
npx reskill@latest install s3://bucket/path/skill.zip

# Registry-based (requires registry deployment)
npx reskill@latest install @scope/skill-name@1.0.0
npx reskill@latest install skill-name

# Install multiple skills at once
npx reskill@latest install github:user/skill1 github:user/skill2@v1.0.0

Monorepo Support

For repositories containing multiple skills, you can install a specific skill by path or install all skills from a parent directory:

bash
# Shorthand format with subpath
npx reskill@latest install github:org/monorepo/skills/planning@v1.0.0
npx reskill@latest install gitlab:company/skills/frontend/components@latest

# URL format with subpath
npx reskill@latest install https://github.com/org/monorepo.git/skills/planning@v1.0.0
npx reskill@latest install git@gitlab.company.com:team/skills.git/backend/apis@v2.0.0

# GitHub web URL automatically extracts subpath
npx reskill@latest install https://github.com/org/monorepo/tree/main/skills/planning

# Point to a parent directory — auto-detects and installs all child skills
npx reskill@latest install https://github.com/org/monorepo/tree/main/skills

When the target directory has no root SKILL.md but contains subdirectories with SKILL.md files, reskill automatically discovers and installs all child skills. Each skill is saved separately in skills.json.

HTTP/OSS URL Support

Skills can be installed directly from HTTP/HTTPS URLs pointing to archive files:

FormatExampleDescription
HTTPS URLhttps://example.com/skill.tar.gzDirect download URL
Aliyun OSShttps://bucket.oss-cn-hangzhou.aliyuncs.com/skill.tar.gzAliyun OSS URL
AWS S3https://bucket.s3.amazonaws.com/skill.tar.gzAWS S3 URL
OSS Protocoloss://bucket/path/skill.tar.gzShorthand for Aliyun OSS
S3 Protocols3://bucket/path/skill.tar.gzShorthand for AWS S3

Supported archive formats: .tar.gz, .tgz, .zip, .tar

Version Formats

FormatExampleDescription
Exact@v1.0.0Lock to specific tag
Latest@latestGet the latest tag
Range@^2.0.0Semver compatible (>=2.0.0 <3.0.0)
Branch@branch:developSpecific branch
Commit@commit:abc1234Specific commit hash
(none)-Default branch (main)

Configuration

skills.json

The project configuration file, created by reskill init:

json
{
  "skills": {
    "planning": "github:user/planning-skill@v1.0.0",
    "internal-tool": "internal:team/tool@latest"
  },
  "registries": {
    "internal": "https://gitlab.company.com"
  },
  "defaults": {
    "installDir": ".skills",
    "targetAgents": ["cursor", "claude-code"],
    "installMode": "symlink"
  }
}
  • skills — Installed skill references (name → source ref)
  • registries — Custom Git registry aliases
  • defaults.installDir — Where skills are stored (default: .skills)
  • defaults.targetAgents — Default agents to install to
  • defaults.installMode — symlink (default, recommended) or copy

Environment Variables

VariableDescriptionDefault
RESKILL_CACHE_DIRGlobal cache directory~/.reskill-cache
RESKILL_TOKENAuth token (takes precedence over ~/.reskillrc)-
RESKILL_REGISTRYDefault registry URLhttps://registry.reskill.dev
RESKILL_NO_MANIFESTSkip skills.json and skills.lock writes (set to 1 to enable)-
DEBUGEnable debug logging-
VERBOSEEnable debug logging (same effect as DEBUG)-
NO_COLORDisable colored output-

Multi-Agent Support

Skills are installed to .skills/ by default and can be integrated with any agent:

AgentPath
Amp.agents/skills
Antigravity.agent/skills
Claude Code.claude/skills
Claude Cowork 3PApp-managed global directory
Clawdbotskills
Codex.codex/skills
Cursor.cursor/skills
Droid.factory/skills
Gemini CLI.gemini/skills
GitHub Copilot.github/skills
Goose.goose/skills
Kilo Code.kilocode/skills
Kiro CLI.kiro/skills
Neovate.neovate/skills
OpenCode.opencode/skills
Roo Code.roo/skills
Trae.trae/skills
Windsurf.windsurf/skills

Use --agent to target specific agents, or --all to install to all detected agents:

bash
# Install to specific agents
reskill install github:user/skill -a cursor claude-code

# Install to all detected agents
reskill install github:user/skill --all

# List skills installed to a specific agent
reskill list -a cursor
reskill list -a claude-cowork-3p

Publishing

Note: Publishing requires a private registry deployment.

Authentication

bash
# Login with a token (obtain from the registry web UI)
reskill login --registry <url> --token <token>

# Check current login status
reskill whoami

# Logout
reskill logout

Tokens are stored in ~/.reskillrc. You can also use the RESKILL_TOKEN environment variable (takes precedence, useful for CI/CD).

Registry URL resolution priority:

  1. --registry CLI option
  2. RESKILL_REGISTRY environment variable
  3. defaults.publishRegistry in skills.json

Publishing a Skill

bash
# Validate without publishing (recommended first step)
reskill publish --dry-run

# Publish the skill
reskill publish

# Publish from a specific directory
reskill publish ./path/to/skill

# Skip confirmation
reskill publish -y

The skill directory must contain a valid SKILL.md with name, version, and description in its frontmatter.

Common Workflows

First-Time Project Setup

bash
# 1. Initialize the project
reskill init -y

# 2. Install skills your project needs
reskill install github:user/skill1@v1.0.0 github:user/skill2@latest -y

# 3. Verify installation
reskill list

# 4. Commit skills.json and skills.lock to version control
# (These files ensure team members get the same skill versions)

Team Collaboration

When a teammate clones the project, they run:

bash
# Reinstall all skills from skills.json (like npm install)
reskill install

This reads skills.json + skills.lock and installs the exact same versions.

Checking and Updating Skills

bash
# Check which skills have newer versions
reskill outdated

# Update all skills
reskill update

# Update a specific skill
reskill update skill-name

Global vs Project-Level Installation

ScopeFlagDirectoryUse Case
Project-.skills/ (in project)Team-shared skills, committed to git
Global-g~/.agents/skills/Personal skills, available in all projects
Claude Cowork 3P-a claude-cowork-3pApp-managed global directoryAlways global, no skills.json writes
bash
# Project-level (default)
reskill install github:user/skill

# Global (personal, all projects)
reskill install github:user/skill -g

# Install to Claude Cowork 3P (always global)
reskill install github:user/skill -a claude-cowork-3p

# Personal project-level (not saved to skills.json)
reskill install github:user/skill --no-save

Diagnosing Issues

bash
# Run environment diagnostics
reskill doctor

# JSON output for programmatic use
reskill doctor --json

The doctor command checks: reskill version, Node.js version, Git availability, cache directory, skills.json validity, skills.lock sync, installed skills integrity, and network connectivity.

Troubleshooting

Error MessageCauseSolution
skills.json not foundProject not initializedRun reskill init
Unknown scope @xyzNo registry configured for this scopeCheck registries in skills.json or use full Git URL
Skill not foundSkill name doesn't exist in registryVerify skill name; check reskill find <query>
Version not foundRequested version doesn't existRun reskill info <skill> to see available versions
Permission deniedAuth issue when publishingRun reskill login; check token scope
Token is invalid or expiredStale authenticationRe-authenticate with reskill login --token <new-token>
Network errorCannot reach Git host or registryCheck network; run reskill doctor for diagnostics
Conflict: directory already existsSkill already installedUse --force to reinstall

Private Repositories

reskill uses your existing git credentials (SSH keys or credential helper). For CI/CD environments:

bash
# GitLab CI
git config --global url."https://gitlab-ci-token:${CI_JOB_TOKEN}@gitlab.company.com/".insteadOf "https://gitlab.company.com/"

Frequently asked questions

What does the Reskill Usage AI skill do?

Teaches AI agents how to use reskill — a Git-based package manager for AI agent skills. Covers CLI commands, install formats, configuration, publishing, and common workflows.

Why use Reskill Usage on TypingMind?

Because you install it once and use it with any model. Reskill Usage is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Reskill Usage in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/kanyun-inc/reskill/tree/main/skills/reskill-usage. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Reskill Usage?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Reskill Usage?

As many as you like. As long as a model supports skills, you can use Reskill Usage with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Reskill Usage AI skill free?

Yes. It is published on GitHub by kanyun-inc under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇