Agent skill

planner

This skill should be used when the user asks to "plan a feature", "create feature plan", "start planning", "new feature", or wants comprehensive planning with analysis, spec, research, design, and TDD task breakdown.

Stars 0
Forks 0

Install this agent skill to your Project

npx add-skill https://github.com/ianphil/my-skills/tree/main/planner

SKILL.md

Feature Planning Session

Create a comprehensive feature planning artifact for: $ARGUMENTS

This planning session follows the Feature Planning Methodology documented in aidocs/feature-planning-methodology.md.

Session Initialization

Step 1: Determine Feature ID

Run this command to find the next feature number:

bash
{ ls backlog/plans/ 2>/dev/null; ls backlog/plans/_completed/ 2>/dev/null; } | grep -E '^[0-9]{3}' | sort -r | head -1

The Feature ID format is {NNN}-{slug} where:

  • NNN = 3-digit sequential number (001, 002, 003...)
  • slug = lowercase, hyphenated descriptor (2-4 words)

Step 2: Create Branch and Folder Structure

bash
git checkout -b feature/{NNN}-{slug}
mkdir -p backlog/plans/{NNN}-{slug}/contracts

Note: When feature is complete, move folder to backlog/plans/_completed/{NNN}-{slug}/

Step 3: Create Planning Artifacts

Create all artifacts in sequence using templates from references/templates.md.


Planning Phases

Phase 1: Analysis (analysis.md)

Goal: Understand the problem space, existing code, and patterns.

Key sections:

  • Executive Summary with pattern/integration mapping
  • Architecture Comparison (current vs target)
  • Pattern Mapping from existing codebase
  • What Exists vs What's Needed
  • Key Insights (what works, gaps/limitations)

Phase 2: Specification (spec.md)

Goal: Define requirements from user perspectives.

Key sections:

  • Problem Statement and Solution Summary
  • User Stories with Acceptance Criteria
  • Functional Requirements (FR-1, FR-2, etc.)
  • Non-Functional Requirements (performance, security)
  • Scope (in/out/future)
  • Success Criteria, Assumptions, Risks

Phase 3: Research (research.md)

Goal: Validate design against external specifications, APIs, or standards.

Key sections:

  • Conformance Analysis against external specs
  • Gap identification
  • Recommendations (critical, minor, future)

Phase 4: Design

Create three artifacts:

4a. Implementation Plan (plan.md)

  • Architecture diagrams (ASCII)
  • Component responsibilities
  • File structure with MODIFY/NEW annotations
  • Key design decisions with rationale
  • Risk mitigation

4b. Data Model (data-model.md)

  • Entity definitions with fields/types
  • Relationships and invariants
  • State transitions
  • Validation rules

4c. Contracts (contracts/)

  • Interface definitions
  • API schemas
  • Configuration schemas

Phase 5: Test Design

5a. Spec Tests (specs/tests/{NNN}-{slug}.md)

CRITICAL: Before writing spec tests, load the spec-tests skill:

Skill(skill="all-skills:spec-tests")

See references/spec-tests-guide.md for detailed writing guidance.

Key principles:

  • Tests verified by reading code, not executing
  • Intent explains WHY users/business care
  • Assertions reference specific files and observable structures

5b. Task Breakdown (tasks.md)

Organize tasks by phase with TDD workflow:

  • [TEST] tasks: Write failing tests first
  • [IMPL] tasks: Implement to pass tests
  • [SPEC] tasks: Run spec tests at phase boundaries

Include dependency diagram and task summary table.


Session Workflow

Execution Order

  1. Initialize: Determine Feature ID, create branch and folders
  2. Phase 1: Create analysis.md - understand problem space
  3. Phase 2: Create spec.md - define requirements
  4. Phase 3: Create research.md - validate against external specs (if applicable)
  5. Phase 4: Create plan.md, data-model.md, contracts/ - design
  6. Phase 5: Create specs/tests/{NNN}-{slug}.md and tasks.md - test design
  7. Cleanup: Remove the corresponding quick plan from backlog/plans/ (e.g., backlog/plans/YYYYMMDD-{slug}.md) since the full planning artifacts supersede it

Interactive Planning

As work progresses through each phase:

  • Read relevant existing code to inform analysis
  • Ask clarifying questions about requirements
  • Research external APIs/specs as needed
  • Validate design decisions with the user

Output Summary

After completing all phases, summarize:

Feature: {NNN}-{slug}
Branch: feature/{NNN}-{slug}

Created Artifacts:
├── backlog/plans/{NNN}-{slug}/
│   ├── analysis.md      ✅
│   ├── spec.md          ✅
│   ├── research.md      ✅
│   ├── plan.md          ✅
│   ├── data-model.md    ✅
│   ├── tasks.md         ✅
│   └── contracts/
│       └── README.md    ✅
└── specs/tests/{NNN}-{slug}.md  ✅

Cleaned Up:
- Removed quick plan: backlog/plans/YYYYMMDD-{slug}.md  🗑️

Next Steps:
1. Review planning artifacts
2. Commit: git add backlog/plans/{NNN}-{slug} specs/tests/{NNN}-{slug}.md && git commit -m "feat: planning for {NNN}-{slug}"
3. Begin implementation: /work-plan {NNN}
4. When complete: mv backlog/plans/{NNN}-{slug} backlog/plans/_completed/

Additional Resources

Reference Files

For detailed templates and guidance, consult:

  • references/templates.md - Full templates for all planning artifacts
  • references/spec-tests-guide.md - Detailed spec test writing guidance

External References

  • Feature Planning Methodology
  • Developer Guide
  • Example: 001-mcp-integration

Expand your agent's capabilities with these related and highly-rated skills.

ianphil/my-skills

commit

This skill should be used when the user asks to "commit changes", "push my code", "commit and push", "save my work", or wants to stage all changes and push to remote.

0 0
Explore
ianphil/my-skills

prime-feat

This skill should be used when the user asks to "load feature context", "resume feature work", "prime feature N", "continue feature N", or wants to load all planning artifacts for a specific feature number.

0 0
Explore
ianphil/my-skills

implement-agents

This skill should be used when the user asks to "implement in parallel", "run phases concurrently", "parallel implement", "implement-agents phase X phase Y", or wants to orchestrate multiple agents running /implement simultaneously.

0 0
Explore
ianphil/my-skills

code-walkthrough

Generate structured, hallucination-proof code walkthroughs using Showboat. Use when the user asks to "walk me through this code", "explain this codebase", "create a walkthrough", "document how this works", "give me a tour of this repo", or wants a readable narrative document that explains code with real verified snippets. Also triggers on "showboat", "linear walkthrough", or "code documentation". Not for quick code questions — this produces a full markdown document.

0 0
Explore
ianphil/my-skills

ainotes

This skill should be used when the user asks to "consolidate notes", "summarize ainotes", "clean up notes", "ainotes", or wants to consolidate accumulated agent observations into a compact summary.

0 0
Explore
ianphil/my-skills

implement

This skill should be used when the user asks to "implement phase N", "implement task T001", "do TDD for phase", "implement T001-T009", or wants to execute tasks following strict TDD red-green-refactor workflow.

0 0
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results