sdd-plan
Core Philosophy
Plan First, Code Second: Every development task begins with a detailed specification that acts as a contract between intent and implementation. This prevents the common failure mode where AI "works once and then falls apart" in real codebases.
Staged Planning (Recommended): For complex features, create specifications in two stages: (1) high-level phase structure for user review and approval, then (2) detailed task breakdown. This reduces wasted effort and enables early course correction before detailed planning begins.
Atomic Tasks: Each task represents a single, focused change to one file. Tasks are the fundamental unit of work in SDD, and keeping them atomic provides:
- Precise dependency tracking: File-level dependencies are explicit and clear
- Granular progress monitoring: Each completed task represents concrete, verifiable progress
- Parallel implementation: Independent tasks can be worked on simultaneously
- Straightforward verification: Each task has a focused scope and clear success criteria
- Easy rollback: Changes can be reverted at the file level without affecting other work
When a feature requires changes across multiple files, decompose it into multiple tasks with proper dependencies, or use subtasks to organize related file changes under a parent task. Never bundle multiple file changes into a single task. Always practice atomic task decomposition and verification.
Key Benefits:
- Reduces hallucinated APIs and misread intent
- Prevents breaking existing functionality
- Provides clear verification criteria
More from tylerburleigh/claude-sdd-toolkit
sdd-next
Task preparation skill for spec-driven workflows. Reads specifications, identifies next actionable tasks, and creates detailed execution plans. Use when ready to implement a task from an existing spec - bridges the gap between planning and coding.
1sdd-pr
AI-powered PR creation after spec completion. Analyzes spec metadata, git diffs, commit history, and journal entries to generate comprehensive PR descriptions with user approval before creation.
1doc-query
Targeted query capabilities for machine-readable codebase documentation with cross-reference tracking, call graph analysis, and workflow automation. Enables fast lookups of classes, functions, dependencies, and function relationships without parsing source code.
1llm-doc-gen
LLM-powered documentation generation for narrative architecture docs, tutorials, and developer guides. Uses AI consultation to create contextual, human-readable documentation from code analysis and spec data.
1sdd-update
Progress tracking for spec-driven development. Use to update task status, track progress, journal decisions, move specs between folders, and maintain spec files. Handles the administrative/clerical aspects of specification documents during development.
1run-tests
Comprehensive pytest testing and debugging framework. Use when running tests, debugging failures, fixing broken tests, or investigating test errors. Includes systematic investigation workflow with external AI tool consultation and verification strategies.
1