Orchestrates the agent pipeline for new features through planning, coding, testing, and review.

Install

mkdir -p .claude/skills/plan-igorilic && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/14319" && unzip -o skill.zip -d .claude/skills/plan-igorilic && rm skill.zip

Installs to .claude/skills/plan-igorilic

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Entry point for all new work. Orchestrates the full agent pipeline: architect (Opus) designs spec+todo, tdd-developer (Sonnet) implements via TDD, qa (Opus) hunts spec/impl gaps, reviewer (Sonnet) reviews code. Triggers on: plan, new feature, start work, build feature.
269 charsno explicit “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)
Intermediate

Key capabilities

  • Architect designs specifications and to-do lists
  • TDD-developer implements features via TDD
  • QA hunts for spec/implementation gaps
  • Reviewer reviews code and presents findings
  • Triage findings into fix, tech debt, or ignore

How it works

The skill orchestrates a multi-agent pipeline where an architect creates specs, a developer implements via TDD, QA verifies, and a reviewer checks the code, with a triage step for findings.

Inputs & outputs

You give it
A description or ticket ID for new work
You get back
A completed feature with reviewed code and triaged findings

When to use plan

  • Plan a new feature
  • Execute development pipeline
  • Audit spec gaps

About this skill

Pipeline: /plan <description or ticket ID>

architect (Opus) → spec.md + todo.md (atomic steps)
  ↓ per step:
tdd-developer (Sonnet) → RED→GREEN→REFACTOR → commit
qa (Opus) → adversarial test + spec-gap & security hunt
reviewer (Sonnet) → present findings → user triages
  ↓ max 3 fix loops, then next step

Usage

  1. /plan <description> → architect creates spec + todo
  2. Approve plan, then: Use tdd-developer on Step 1 of <todo>.md
  3. After each step: Use qa to verifyUse reviewer to review
  4. Triage: [F]ix / [T]ech debt / [I]gnore → next step

When not to use it

  • When a single, atomic task needs to be executed without a full pipeline

Limitations

  • Requires manual approval of the plan
  • Requires manual invocation of subsequent steps like `tdd-developer`, `qa`, and `reviewer`
  • Limited to a maximum of 3 fix loops before moving to the next step

How it compares

This skill automates and structures the entire development workflow through a series of specialized agents, providing a guided process compared to a manual, less coordinated approach.

Compared to similar skills

plan side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
plan (this skill)01moNo flagsIntermediate
brainstorming934moReviewBeginner
spec-workflow108moNo flagsIntermediate
feature-design-assistant67moNo flagsIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry