code-task-generator

What This Skill Does

The code-task-generator is a specialized developer tool designed to bridge the gap between high-level architectural planning and actionable, granular engineering tasks. By parsing either loose natural language descriptions or structured PDD (Plan-Driven Development) documents, the skill automatically identifies scope, technical requirements, and functional dependencies. It transforms these inputs into standardized .code-task.md files, ensuring that developers have clear, consistent blueprints for implementation that include baked-in Gherkin-style (Given-When-Then) acceptance criteria.

Installation

You can integrate this skill into your OpenClaw environment by running the following command in your terminal: clawhub install openclaw/skills/skills/paulpete/code-task-generator

Use Cases

• PDD Execution: Follow complex, multi-step implementation plans defined in markdown, ensuring stateful progression between steps while maintaining context.
• Feature Scoping: Convert vague product requirements or brainstormed ideas into structured development tasks, preventing scope creep and ensuring test coverage is planned from the start.
• Standardization: Enforce a consistent format for task definitions across your engineering team, making it easier for AI agents or junior developers to understand requirements and constraints.
• Project Onboarding: Quickly generate a roadmap and granular task list from a directory of project research documents or architectural design files.

Example Prompts

1. "I have a plan.md file in the current directory. Please use the code-task-generator to analyze it and propose a breakdown for step 2."
2. "Generate a code task for building a user authentication endpoint using FastAPI. Constraints: must use OAuth2, handle JWT tokens, and include unit tests for token validation."
3. "Review the design document at ./docs/arch/auth-system.md and create a task list for implementing the Redis caching layer."

Tips & Limitations

• Approval Workflow: Always review the task breakdown presented by the AI before confirming file generation. This is a safety feature to ensure the agent's logic aligns with your architectural vision.
• PDD Context: When using PDD mode, ensure your plan is formatted with clear steps; the skill relies on this structure to accurately track your progress through the implementation cycle.
• Test Integration: Remember that this skill integrates testing requirements directly into tasks. You do not need to create separate 'write tests' tickets—the agent will ensure they are part of the acceptance criteria from the start.
• Scope: While the tool is powerful, it is limited by the quality of the input. Vague descriptions will yield less precise acceptance criteria than well-defined technical specifications.