Decomposes features into actionable implementation plans using parallel worker-explorer swarms. Supports decision frameworks, artifact generation, and task tracking.
How this skill is triggered — by the user, by Claude, or both
Slash command
/agentic-framework:swarm-planThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Decompose features into actionable plans using parallel exploration swarms.
Decompose features into actionable plans using parallel exploration swarms.
Sequential Thinking (structured analysis):
Context7 (library research):
| Decision Type | Reversibility | Required Artifacts |
|---|---|---|
| Two-Way Door | Easy to reverse | PR description only |
| One-Way Door (Medium) | Moderate effort | RFC + Design excerpt |
| One-Way Door (High) | Expensive/impossible | Full ADR + Stakeholder review |
Small Feature (1-3 days)
plan_[feature].md — Implementation steps onlyMedium Feature (1-2 weeks)
prd_[feature].md — Requirementsplan_[feature].md — Implementation stepsLarge Feature (2+ weeks)
pr_faq_[feature].md — Vision and customer valueprd_[feature].md — Detailed requirementsadr_[key-decision].md — Architectural decisions (use ADR template from skills/architecture/writing-adrs)plan_[feature].md — Implementation steps| Worker | Primary Use |
|---|---|
worker-explorer | Fast codebase search, web research, dependency mapping |
worker-builder | Implementation, testing, refactoring |
worker-reviewer | Code review, security audit, quality assessment |
worker-research | Deep multi-source investigation, technology evaluation |
worker-architect | Complex design decisions, ADRs, system architecture |
Model tiers are pinned in each agent's frontmatter (.claude/agents/) — that is the single source of truth.
Orchestrator spawns 4-8 worker-explorer agents simultaneously
Each searches different parts of codebase
Results aggregated for next phase
1. worker-architect designs solution
2. Orchestrator decomposes into N tasks
3. N worker-builder agents execute in parallel
4. worker-reviewer validates each output
5. Orchestrator integrates
worker-reviewer (focus: security) scans all components in parallel
Findings aggregated and prioritized
worker-builder fixes critical/high issues
# Launch exploration workers in parallel via Task tool
# Each worker focuses on one aspect:
# - Existing patterns in codebase
# - External dependencies and APIs
# - Security and performance constraints
# - Related ADRs and design specs
Use TaskCreate to record implementation tasks, each with clear acceptance criteria:
TaskCreate — "Implement [feature]" (epic-level task)TaskCreate — "[Task 1: Foundation]" with acceptance criteriaTaskCreate — "[Task 2: Core Logic]" with acceptance criteriaLink dependencies with TaskUpdate (addBlockedBy): Task 2 gets Task 1 added to its blockedBy list so it can't start until Task 1 completes.
./artifacts/Every planning session MUST produce:
./artifacts/ following naming conventionsTaskCreate) for all implementation work, each with acceptance criteriablockedBy)Use writing-pr-faqs skill for structure and template.
Use writing-prds skill for structure and template.
swarm-coordination, writing-adrs, designing-systems, designing-apis, writing-pr-faqs, writing-prds
/swarm-execute: Plan artifact + task list ready/architect: Complex decisions requiring ADR reviewnpx claudepluginhub dralgorhythm/claude-agentic-frameworkCreates dependency-aware implementation plans optimized for parallel multi-agent execution. Useful for coordinating complex feature implementation across multiple agents.
Conducts Socratic planning interview to decompose tasks into independent, testable stories for parallel swarm execution. Explores codebase, decisions, and context first.