Agent skill
design-plan
Create implementation plans for complex tasks with architectural guidance. Use for complex features, multi-file changes, or when planning is needed before implementation.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/design-plan
SKILL.md
Design Plan
Purpose
Create comprehensive implementation plans for complex tasks, providing architectural guidance that can be followed for implementation.
When to Use
- Complex features with architectural decisions
- Tasks spanning multiple files/modules
- New patterns or abstractions needed
- When user requests
--with-architectflag
Task-Specific Rules
Based on task type, reference these rules when planning:
| Task Type | Rules to Reference |
|---|---|
| Component | components.md, figma.md |
| API/Backend | context-engine.md |
| Token System | tokens.md |
Always also load: Base rules (workflow, github, linear)
Planning Process
Phase 1: Understand Requirements
-
Read the Linear ticket thoroughly:
mcp__linear__get_issue(id: "{issue_id}", includeRelations: true) -
Extract key information:
- What problem is being solved?
- What are the acceptance criteria?
- Any constraints or requirements?
- Linked designs or references?
-
If Figma link present:
- Use
mcp__figma__get_design_contextto understand the design - Note complexity and variants
- Use
Phase 2: Research Technology Context
IMPORTANT: Before designing, understand the current state of technologies involved.
-
Identify technologies in scope:
- What libraries/frameworks will be used?
- What version is the project using? (check
package.json)
-
Use WebSearch to research:
- Latest best practices for the technology
- Version-specific features or limitations
- Known issues, deprecations, or breaking changes
- Recommended patterns from official docs
-
Key questions to answer:
Question Why It Matters What version are we using? APIs differ between versions What's the recommended pattern? Avoid deprecated approaches Any known issues? Prevent predictable problems What's new in this version? Leverage latest features -
Document findings that affect the design decision
Phase 3: Explore Existing Architecture
-
Find related code:
- How do similar features work?
- What patterns are established?
- What can be reused?
-
Identify integration points:
- What existing modules will this touch?
- What APIs/interfaces exist?
- What dependencies are involved?
-
Note constraints:
- Performance requirements
- Backwards compatibility
- Security considerations
Phase 4: Design the Solution
-
Consider multiple approaches:
- What are the options?
- What are the trade-offs?
-
Choose the best approach based on:
- Alignment with existing patterns
- Maintainability
- Scalability
- Simplicity (avoid over-engineering)
-
Define the architecture:
- File structure
- Data flow
- Component/module boundaries
- Interfaces between parts
Phase 5: Create Implementation Plan
-
Break down into phases:
- Each phase should be independently testable
- Order by dependencies (foundation first)
- Include testing in each phase
-
For each phase, specify:
- What files to create/modify
- What the code should do
- What patterns to follow
- What to test
-
Identify risks and decisions:
- What needs user input?
- What could go wrong?
- What assumptions are being made?
Phase 6: Present Plan for Approval
Output the plan and WAIT for user approval before implementation.
Output Format
## 🏛️ Implementation Plan
### Task
- **Linear:** {issue_id} - {title}
- **Complexity:** {Low|Medium|High}
- **Estimated Phases:** {count}
### Overview
{1-2 sentence summary of what will be built and how}
### Architecture Decision
**Approach:** {chosen approach}
**Why this approach:**
- {reason 1}
- {reason 2}
**Alternatives considered:**
| Alternative | Why Not |
|-------------|---------|
| {option} | {reason} |
### File Structure
{path}/ ├── {file1} — {purpose} ├── {file2} — {purpose} └── {file3} — {purpose}
### Implementation Phases
#### Phase 1: {name}
**Goal:** {what this phase accomplishes}
**Files:**
- Create `{path/file}` — {description}
- Modify `{path/file}` — {description}
**Key Implementation Notes:**
- {note 1}
- {note 2}
**Verification:**
- {how to verify this phase works}
---
#### Phase 2: {name}
{same structure}
---
### Risks & Decisions
| Item | Type | Notes |
|------|------|-------|
| {item} | Risk/Decision | {details} |
### Questions for User
{Any clarifications needed before proceeding}
---
**Ready for implementation?** Approve to proceed.
Principles to Follow
- Design for the codebase — Follow existing patterns, don't invent new ones
- Simple over clever — The best architecture is the simplest one that works
- Think in phases — Break work into testable increments
- Surface decisions — Don't hide assumptions, make them explicit
- Plan for handoff — Implementation should be possible without guessing
What NOT to Do
- Don't over-engineer for hypothetical requirements
- Don't introduce new patterns when existing ones work
- Don't skip the trade-off analysis
- Don't assume — ask user when unclear
- Don't write code — only plan (implementation is separate)
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?