Agentpack is a control plane for deploying coding-agent assets across tools. It manages agent instructions and skills, benefiting developers and operations teams. It integrates with CLI tools and supports Claude agents.
git clone https://github.com/liqiongyu/agentpack.gitAgentpack is a declarative control plane that manages and deploys coding-agent assets—including agent instructions, skills, Claude Code slash commands, and Codex custom prompts—across multiple tools from a centralized manifest. It enables cross-tool consistency by treating your agent configuration as a single source of truth, synced via git and deployable with preview-first, diff-aware workflows. The tool provides snapshot-based rollback capabilities and manifest-protected safe deletes, making deployments reusable and auditable. Developers and operations teams use Agentpack to maintain consistent agent setups across machines and tools, with stable automation contracts via JSON and MCP for orchestration.
Install Agentpack via cargo or prebuilt binaries, then run `agentpack init` to create a manifest. Use `agentpack update` to refresh, `agentpack preview --diff` to review changes, and `agentpack deploy --apply` to deploy across your configured tools. For automation, use `--json` and `--yes` flags with `preview` before applying.
Deploy the same agent instructions and skills to Codex, Claude Code, Cursor, and VSCode simultaneously
Preview and diff agent asset changes before applying deployments across tools
Sync agent configurations across multiple machines using git as a single source of truth
Rollback agent deployments to previous snapshots with full audit trails
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/liqiongyu/agentpackCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
I want to use Agentpack to deploy a coding agent for [COMPANY] in the [INDUSTRY] sector. The agent should be able to [SPECIFIC TASK]. Provide a step-by-step guide to set this up, including any necessary configurations and best practices.
# Agentpack Deployment Guide for [COMPANY] in [INDUSTRY] ## Prerequisites - Agentpack CLI installed - Access to [COMPANY] development environment - Clear understanding of [SPECIFIC TASK] ## Step 1: Initialize Agentpack ```bash agentpack init ``` ## Step 2: Define Agent Instructions Create an `AGENTS.md` file with the following content: ```markdown # Agent Instructions - **Objective**: [SPECIFIC TASK] - **Scope**: [INDUSTRY] sector - **Constraints**: [ANY CONSTRAINTS] ``` ## Step 3: Deploy the Agent ```bash agentpack deploy --agent AGENTS.md ``` ## Step 4: Monitor and Optimize - Use Agentpack's monitoring tools to track performance - Adjust configurations as needed based on initial results
Cloud ETL platform for non-technical data integration
IronCalc is a spreadsheet engine and ecosystem
Get more done every day with Microsoft Teams – powered by AI
Customer feedback management made simple
Enterprise workflow automation and service management platform
Automate your spreadsheet tasks with AI power
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan