task-decomposition
Breaks features into atomic, independently verifiable tasks. No task should take more than 4 hours. Unblocks parallel work and reduces integration risk.
git clone --depth 1 https://github.com/DevelopersGlobal/ai-agent-skills /tmp/task-decomposition && cp -r /tmp/task-decomposition/skills/task-decomposition ~/.claude/skills/task-decompositionSKILL.md
## Overview Large tasks fail in large ways. Decomposed tasks fail in small, fixable ways. This skill breaks any feature or project into atomic tasks — each independently implementable, testable, and deployable. ## When to Use - Before starting any feature that takes more than half a day - When a task feels overwhelming or unclear - When multiple people need to work in parallel ## Process ### Step 1: Identify the Deliverable 1. State what "done" looks like for the whole feature. 2. List all the things that must be true when it's complete. 3. Identify dependencies: what must exist before any task can start? **Verify:** You can state the full feature goal in 2 sentences. ### Step 2: Decompose Into Atomic Tasks 4. Break the feature into tasks where each task: - Can be completed in under 4 hours - Has a single, clear output - Can be verified independently - Can be reverted without breaking other tasks 5. Each task should be: `[verb] [noun] so that [outcome]` - ✅ "Add rate limiting to /api/login so that brute-force is prevented" - ❌ "Work on the login security stuff" **Verify:** Every task is under 4 hours. Every task has a clear verify condition. ### Step 3: Order and Parallelism 6. Draw the dependency graph — what blocks what? 7. Identify tasks that can be done in parallel. 8. Sequence tasks so integration happens incrementally (not as one big bang at the end). **Verify:** The dependency order is clear. No unnecessary sequential dependencies. ### Step 4: Estimate and Adjust 9. For each task, estimate: best case / worst case / expected. 10. If any task's worst case > 1 day: decompose further. 11. Total estimate sanity check: does it add up to a reasonable timeline? ## Common Rationalizations (and Rebuttals) | Excuse | Rebuttal | |--------|----------| | "I'll figure it out as I go" | Decomposition takes 30 minutes. Rework from poor planning takes days. | | "It's too complex to break down" | Everything can be decomposed. Start with what you know, decompose the rest later. | | "The tasks are too granular" | Granular tasks ship continuously. Coarse tasks ship never. | ## Verification - [ ] Each task is under 4 hours - [ ] Each task has a single verify condition - [ ] Dependencies mapped — no circular dependencies - [ ] Parallel work identified ## References - [goal-driven-execution skill](../goal-driven-execution/SKILL.md) - [multi-agent-orchestration skill](../multi-agent-orchestration/SKILL.md)
Validates, parses, and sanitizes AI-generated outputs before they reach end users or downstream systems. Structured output enforcement, schema validation, and fallback handling.
Design stable, versioned, self-documenting APIs. Easy to use correctly, hard to use incorrectly. Apply Hyrum's Law from day one.
Automated quality gates from commit to production. Every merge to main is potentially shippable. No manual steps in the deployment path.
Get layered, context-aware explanations of unfamiliar code. Understand what it does, why it was written that way, and how to work with it safely.
Structured code review focusing on correctness, security, and maintainability. Correctness before style. Every reviewer comment must be actionable.