Miyabi is an open-source, economically-governed, beginner-friendly autonomous development framework built on Issue-Driven Development. It enables operations teams to automate software development tasks, reducing manual effort and accelerating project delivery. Miyabi integrates with GitHub Actions and supports parallel execution, making it ideal for teams looking to streamline their development workflows.
git clone https://github.com/ShunsukeHayashi/Miyabi.gitMiyabi is an open-source autonomous development framework designed to automate software development tasks within operations teams. Built on Issue-Driven Development principles and governed by economic mechanisms, it enables beginner-friendly automation without requiring deep expertise. The framework integrates directly with GitHub Actions and supports parallel task execution, allowing teams to reduce manual effort and accelerate project delivery through streamlined development workflows.
1. **Set Up Miyabi:** Install Miyabi in your GitHub repository using the official GitHub Action (`miyabi/miyabi-action@v1`). Configure the `.miyabi/config.yml` file to define your automation pipelines and parallel execution rules. 2. **Define Automation Tasks:** Create GitHub issues with specific labels (e.g., `automation:dependency-update`) that trigger Miyabi workflows. Use the `miyabi/issue-template.yml` to standardize issue descriptions for consistent automation. 3. **Monitor Execution:** Track progress in the Miyabi dashboard (accessible via GitHub Actions logs or the Miyabi web interface). Use the `miyabi/status` command to check real-time execution status and resource usage. 4. **Review Results:** After execution, review the generated pull requests, logs, and reports. Use Miyabi's `miyabi/summarize` command to generate a concise report of key metrics and recommendations for future optimizations. 5. **Iterate and Improve:** Analyze execution metrics (runtime, resource usage) to identify bottlenecks. Adjust parallel execution settings or add pre/post-execution hooks to improve efficiency. Share successful configurations across teams via Miyabi's shared pipeline library.
Automate repetitive GitHub-based development tasks and workflows
Execute multiple development operations in parallel to reduce project timelines
Enable operations teams to manage software development automation with minimal manual intervention
Integrate issue-driven development practices with CI/CD pipelines via GitHub Actions
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/ShunsukeHayashi/MiyabiCopy 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.
Use Miyabi to automate [SPECIFIC_TASK] in [PROJECT_NAME] repository. Follow the Issue-Driven Development workflow: create a GitHub issue with label '[LABEL]' describing the task, assign it to the Miyabi automation pipeline, and monitor progress in the Miyabi dashboard. Ensure parallel execution is enabled for [SPECIFIC_STEPS] to optimize performance. Provide a summary of the automation results including execution time, resource usage, and any errors encountered.
### Miyabi Automation Report for Project 'E-Commerce Platform' **Task Automated:** Dependency Update for `react` (v18.2.0 → v18.3.0) **Workflow Execution:** 1. **Issue Creation:** GitHub Issue #423 created with label `automation:dependency-update` and title 'Update React to v18.3.0'. Assigned to Miyabi pipeline via GitHub Actions. 2. **Parallel Execution:** Miyabi split the task into 3 parallel sub-tasks: - Sub-task 1: Update `package.json` and `yarn.lock` files - Sub-task 2: Run `yarn audit` to check for vulnerabilities - Sub-task 3: Execute test suite for affected components 3. **Execution Metrics:** - Total runtime: 8 minutes 42 seconds (vs. estimated 15 minutes for sequential execution) - CPU usage: 42% average across 3 parallel workers - Memory usage: 1.2GB peak - Sub-task 2 completed with 1 low-severity vulnerability detected (patched automatically) 4. **Results:** - All dependency files updated successfully - Test suite passed (1234/1234 tests) - Pull request #424 created and merged automatically - No manual intervention required **Recommendations:** - Future dependency updates can reuse this pipeline configuration - Consider adding a pre-execution hook to run `npm outdated` for proactive updates - Monitor GitHub Actions logs for any timeout issues during peak hours
Create and collaborate on interactive animations with powerful, user-friendly tools.
Coordinated robotic swarms for manufacturing
AI-powered financial advisor and portfolio manager
Get more done every day with Microsoft Teams – powered by AI
IronCalc is a spreadsheet engine and ecosystem
Customer feedback management made simple
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan