OpenPackage

🥈Silver

OpenPackage is an open-source alternative to Claude Code Plugins. It enables developers to create, manage, and deploy AI agents with custom skills. Operations teams benefit from streamlined workflows and improved automation. It connects to CLI tools and integrates with existing development environments.

505140Updated 2w ago

Intermediate30min to implementautomation

Saves ~90 min per use

Quick InstallView Source

git clone https://github.com/enulus/OpenPackage.git

Works with:

Claude

Overview

About This Skill

How to Use

1. **Define the Task**: Specify the goal (e.g., 'Deploy a monitoring agent') and list required skills (e.g., `docker`, `prometheus`). 2. **Configure Environment**: Provide the target environment (e.g., `k8s-cluster`, `local-machine`) and CLI tools (e.g., `kubectl`, `docker-compose`). 3. **Set Constraints**: Define limits (e.g., 'Max 2 retries', 'Timeout: 5 minutes') to avoid runaway processes. 4. **Run the Agent**: Execute the agent via `op openpackage --task "[TASK]" --config "[CONFIG_FILE]"` or integrate it into a CI/CD pipeline. 5. **Review Outputs**: Check the agent's real-time logs and final report for actions taken, errors, and recommendations. Use `--verbose` for detailed logs. **Tips**: - Start with a small, testable task to validate the agent's behavior. - Use `op validate` to pre-check configurations before execution. - Store common configurations in YAML files (e.g., `configs/ci-pipeline.yaml`) for reuse.

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/enulus/OpenPackage

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Act as an OpenPackage agent to [TASK]. Use the following configuration: 
- Skills: [LIST_OF_SKILLS]
- Environment: [ENVIRONMENT_NAME]
- CLI Tools: [TOOLS_IF_ANY]
- Constraints: [ANY_CONSTRAINTS]

Start by validating the configuration, then execute the task step-by-step. Report progress in real-time and provide a summary of actions taken, outputs generated, and next steps. If errors occur, diagnose and suggest fixes.

Example Output

### OpenPackage Agent Report
**Task**: Deploy a CI/CD pipeline agent to automate testing for a Python project.

**Configuration Validation**:
- Skills: `python-test-runner`, `git-clone`, `docker-build`
- Environment: `dev-container`
- CLI Tools: `docker`, `git`, `pytest`
- Constraints: Max 3 concurrent jobs, timeout after 10 minutes

**Execution Steps**:
1. **Environment Setup**: Detected `dev-container` with Python 3.11 and Docker installed. No conflicts found.
2. **Repository Cloning**: Cloned `https://github.com/example/repo` into `/workspace/repo`. Branch: `main`. SHA: `a1b2c3d4`.
3. **Dependency Installation**: Installed `pytest`, `pytest-cov`, and `requests` via pip. Logs show 12 packages installed.
4. **Test Execution**: Ran `pytest --cov=src` in `/workspace/repo`. Results:
   - Total tests: 42
   - Passed: 40 (95.2%)
   - Failed: 2 (4.8%)
   - Coverage: 87%
5. **Artifact Generation**: Built Docker image `example/repo:latest` with size 1.2GB. Push skipped due to `--no-push` flag.

**Errors Encountered**:
- Test `test_api_integration` failed due to missing API key. Suggested fix: Add `API_KEY` to environment variables or mock the API.

**Summary**:
- Tasks completed: 4/5 (80%)
- Outputs generated: Test report (`/workspace/repo/test-results.xml`), Docker image (`example/repo:latest`)
- Next steps: Fix failing tests, push Docker image to registry, or trigger downstream jobs.

**Recommendations**:
- Add `API_KEY` to secrets management for future runs.
- Increase timeout to 15 minutes for larger projects.
- Log test results to a persistent storage (e.g., S3) for historical analysis.