Banana Skills is a collection of AI agent skills for Claude Code and other AI coding assistants. It helps developers automate coding tasks, recommend relevant skills, and improve productivity. The skills integrate with AI coding assistants to streamline workflows and enhance coding efficiency.
git clone https://github.com/TreyDong/banana-skills.gitBanana Skills is a collection of three specialized AI agent skills designed to streamline coding workflows. The banana-skill-finder automatically recommends relevant Claude skills based on your tasks, using a three-tier search strategy across SkillsMP, skills.sh, and GitHub. banana-sync-to-notion enables developers to sync local Markdown files to Notion with full formatting preservation and automatic duplicate detection. claude-codex-import lets you migrate Claude Code and Codex conversation history into memory systems for future reference. These skills work with Claude Code, Cursor, Windsurf, GitHub Copilot, and other AI assistants supporting the SKILL.md standard, helping teams automate repetitive tasks and improve documentation workflows.
1. **Identify the Task:** Clearly define the coding task you want to automate (e.g., refactoring, testing, or dependency updates). Use specific language like 'Refactor this JavaScript class to use ES6 modules' or 'Add unit tests to this Python module'. 2. **Select Banana Skills:** Run `banana list-skills` in your terminal to view available skills. Choose the ones most relevant to your task (e.g., `python-type-hint-generator` for Python projects). 3. **Execute Automation:** Use the `banana apply` command with your selected skills. For example: ```bash banana apply python-type-hint-generator pep8-formatter --project-path ./src ``` Replace `[TASK_DESCRIPTION]` and `[PROJECT_PATH]` in the prompt template with your specific details. 4. **Review Output:** Check the generated report for changes, errors, and recommendations. Use the `banana validate` command to test the automated changes (e.g., `banana validate --test-suite pytest`). 5. **Iterate if Needed:** If the automation misses edge cases, refine your task description or add specific skills. For example, use `banana apply legacy-code-optimizer --custom-rules ./rules.json` to apply custom rules.
Automatically discover and install skills when encountering specialized coding tasks
Back up Markdown documentation to Notion with formatting and structure preserved
Import Claude Code and Codex chat history for building agent memory systems
Migrate coding conversations into searchable knowledge archives
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/TreyDong/banana-skillsCopy 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 Banana Skills to automate the following coding task: [TASK_DESCRIPTION]. Identify the most relevant Banana Skills for this task from the available collection. Apply the recommended skills to [PROJECT_PATH] and execute the automation. Generate a report summarizing the changes made, code improvements, and any errors encountered during execution.
### Banana Skills Automation Report
**Task Executed:** Automated refactoring of legacy Python code in `/projects/legacy_system` to modernize it with type hints and PEP 8 compliance.
**Recommended Skills Applied:**
1. **`python-type-hint-generator`**: Added type hints to all functions and class methods. For example:
```python
# Before
def process_data(data):
...
# After
def process_data(data: list[dict]) -> tuple[dict, int]:
...
```
2. **`pep8-formatter`**: Reformatted all files to adhere to PEP 8 standards. Fixed issues like:
- Line lengths exceeding 79 characters
- Inconsistent indentation
- Missing docstrings
3. **`legacy-code-optimizer`**: Identified and replaced deprecated patterns (e.g., `map()` with list comprehensions) and optimized nested loops.
**Changes Summary:**
- **Files Modified:** 12
- **Lines Added:** 450 (type hints, docstrings)
- **Lines Removed:** 180 (redundant or deprecated code)
- **Errors Encountered:** 3 (resolved via `error-handler` skill)
**Next Steps:**
- Run `pytest` to validate the refactored code.
- Merge changes into the `dev` branch after review.
**Performance Impact:** Reduced cyclomatic complexity by 22% and improved IDE linting score from 6.5 to 9.2.
AI assistant built for thoughtful, nuanced conversation
IronCalc is a spreadsheet engine and ecosystem
ITIL-aligned IT service management platform
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