1. Conduid
  2. Agents
  3. Agr
MCP server · Agents

Agr

A package manager for AI agents. Install agent skills from GitHub with a single command.

52Fair

Scored 5 months ago · breakdown

About Agr

Agr is an MCP server in the Agents category: a package manager for AI agents. Install agent skills from GitHub with a single command. It has been installed 0 times through Conduid.

Install

Clone
git clone https://github.com/computerlovetech/agr

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

Ask AI about Agr

Powered by Claude · Grounded in docs

I know everything about Agr. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

  • ·README presentNot checked yet.
  • ·License declaredNot checked yet.
  • ·Tests presentNot checked yet.
  • ·Dependencies pinnedNot checked yet.
  • ·No dynamic code executionNot checked yet.
  • ·Scoped permissionsNot checked yet.

README

agr

The package manager for AI agents.

Install, share, and sync agent resources across Claude Code, Cursor, Codex, OpenCode, Copilot, and Antigravity — built for teams practicing Agentic Engineering.

PyPI Docs


Getting Started

Install the CLI:

uv tool install agr

Install your first skill:

agr add anthropics/skills/frontend-design

Then invoke it in your AI tool:

Tool Invoke with
Claude Code /frontend-design
Cursor /frontend-design
OpenAI Codex $frontend-design
OpenCode frontend-design
GitHub Copilot /frontend-design
Antigravity (via IDE)

No setup required — agr add auto-creates agr.toml and detects which tools you use.


Run a skill without installing

agrx downloads a skill, runs it with your tool's CLI, and cleans up. Nothing is saved to your project:

agrx anthropics/skills/pdf -p "Extract tables from report.pdf"
agrx anthropics/skills/skill-creator -i   # Interactive: skill + chat

Built for teams

agr is built for teams practicing Agentic Engineering — where AI agents are first-class members of your development workflow. Share a consistent set of agent resources across every developer and every AI tool on your team.

Dependencies are tracked in agr.toml — commit it, and teammates install everything with one command:

dependencies = [
    {handle = "anthropics/skills/frontend-design", type = "skill"},
    {handle = "anthropics/skills/pdf", type = "skill"},
    {handle = "your-team/agent-resources/dev-workflow", type = "package"},
]
agr sync   # Like npm install, but for AI agents

New teammate? agr sync and they're productive on day one — same agents, same skills, same standards.


Create and share

agr init my-skill                # Scaffold a new skill
# Edit my-skill/SKILL.md with your instructions
agr add ./my-skill               # Test locally
# Push to GitHub, then others can:
agr add your-username/my-skill

Python SDK

Use agr as a library to load, discover, and cache skills programmatically:

from agr import Skill, list_skills, skill_info

skill = Skill.from_git("anthropics/skills/code-review")
print(skill.prompt)   # SKILL.md contents
print(skill.files)    # Files in the skill directory

for info in list_skills("anthropics/skills"):
    details = skill_info(info.handle)
    print(f"{details.handle}: {details.description}")

See the SDK documentation for cache management, error handling, and the full API.


Example skills

Documents & data:

agr add anthropics/skills/pdf              # Read, extract, create PDFs
agr add anthropics/skills/docx             # Generate and edit Word documents
agr add anthropics/skills/xlsx             # Build and manipulate spreadsheets
agr add anthropics/skills/pptx             # Create and work with slide decks
agr add anthropics/skills/doc-coauthoring  # Structured doc co-authoring workflow

Design & frontend:

agr add anthropics/skills/frontend-design   # Production-grade interfaces
agr add anthropics/skills/canvas-design     # Visual art in PNG and PDF
agr add anthropics/skills/algorithmic-art   # Algorithmic art with p5.js
agr add anthropics/skills/theme-factory     # Style artifacts with themes
agr add anthropics/skills/brand-guidelines  # Anthropic brand colors and typography

Development:

agr add anthropics/skills/claude-api             # Build apps with the Claude API
agr add anthropics/skills/mcp-builder            # Create MCP servers
agr add anthropics/skills/web-artifacts-builder  # Multi-component HTML artifacts
agr add anthropics/skills/webapp-testing         # Test web apps with Playwright

Productivity:

agr add anthropics/skills/skill-creator     # Create, modify, and improve skills
agr add anthropics/skills/internal-comms    # Write internal communications
agr add anthropics/skills/slack-gif-creator # Create animated GIFs for Slack

Browse the full list in the Skill Directory.


All commands

Command Description
agr add <handle> Install a skill
agr add <handle> -o Update to the latest version
agr remove <handle> Uninstall a skill
agr sync Install all from agr.toml
agr upgrade [handle...] Re-fetch deps past the pinned commit
agr list Show installed skills
agr init Create agr.toml (auto-detects tools)
agr init <name> Create a new skill
agr config <cmd> Manage configuration
agrx <handle> Run a skill temporarily

Add -g to add, remove, sync, or list for global skills (available in all projects).


Community skills

# Go — @dsjacobsen
agr add dsjacobsen/agent-resources/golang-pro

# Drupal & DevOps — @madsnorgaard
agr add madsnorgaard/drupal-agent-resources/drupal-expert
agr add madsnorgaard/drupal-agent-resources/drupal-security
agr add madsnorgaard/drupal-agent-resources/drupal-migration
agr add madsnorgaard/drupal-agent-resources/ddev-expert
agr add madsnorgaard/drupal-agent-resources/docker-local

# Workflow — @maragudk
agr add maragudk/skills/collaboration

Browse all community skills in the Skill Directory.

Built something? Share it here.


Documentation · Skill Directory · Tutorial · Python SDK · MIT License

README mirrored from the source repository 5 months ago. The original is authoritative.

Questions

About Agr

How do I install Agr?

Run git clone https://github.com/computerlovetech/agr, then add the server to your MCP client's configuration. Conduid has recorded 0 installs, so the command is known to work with current clients.

Is Agr safe to use with an AI agent?

Its trust score is 52 out of 100 (fair). Conduid hasn't run static security checks on this repository yet, so review the source yourself before granting it credentials. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Agr still maintained?

Conduid hasn't recorded a commit date for this repository yet. Check the repository directly for recent activity.