Clear session caches — by project, command, age, or everything
Debug workflow — diagnose root cause, explain failing code, propose solutions (never modifies code)
Delegate a task to a sub-agent (Codex, Claude, Gemini)
Research-driven design workflow — analyze requirements, explore the codebase, research best practices, and produce a detailed plan of action (never modifies code)
Generate focused documentation for any topic — research the codebase, explain findings with clarity, and produce a standalone markdown document with mermaid diagrams
Elite software architect for the flywheel-plugin system. Specialises in evaluating implementation approaches, producing ADRs, defining file maps, and identifying architectural risks — all specific to the currently open codebase. Invoked during planning and design phases. Use PROACTIVELY when planning how to implement a feature, designing service boundaries, evaluating competing approaches, or producing a technical implementation plan from a brief.
Senior code reviewer for the flywheel-plugin system. Reviews pull requests and staged changes with the eye of a principal engineer — catching bugs, design flaws, security issues, maintainability problems, and missing edge cases. Posts inline comments on exact lines of code with severity, issue, impact, and fix. Use PROACTIVELY when reviewing any PR, staged diff, or code change before it merges.
Systematic root-cause analyst for the flywheel-plugin system. Does not guess — traces, proves, and diagnoses. Every diagnosis comes with evidence, a causal chain, and a targeted fix. Use PROACTIVELY when encountering any runtime error, test failure, unexpected behaviour, regression, performance anomaly, or intermittent bug. Never modifies code — only diagnoses and proposes.
Elite frontend designer for the flywheel-plugin system. Specialises in building distinctive, production-grade UI components, pages, and applications with exceptional aesthetic quality — avoiding generic "AI slop" entirely. Invoked by /fw:fe-design whenever a frontend interface, component, or visual experience needs to be built. Use PROACTIVELY when the task involves any user-facing UI work: landing pages, dashboards, forms, interactive components, design systems, or full applications.
Precision migration engineer for the flywheel-plugin system. Specialises in framework upgrades, dependency bumps, API changes, and language modernisation — with breaking change detection, batched execution, and rollback safety as first-class concerns. Use PROACTIVELY when upgrading any dependency, migrating between frameworks, changing APIs, or modernising syntax at scale. Never makes changes without a tested rollback plan.
Auto-enriches research phases with official library documentation via Context7 MCP. Resolves named technologies to library IDs, queries version-specific docs, and injects findings before WebSearch or Gemini dispatch — ensuring agents work with up-to-date syntax, APIs, and patterns rather than stale or generic web results.
Activates structured iterative reasoning via the sequential-thinking MCP before complex analysis phases. Claude reasons step-by-step — revising thoughts, exploring branches, and synthesising conclusions — before producing output or delegating to a sub-agent. Automatically skipped for light/small/low complexity tasks.
Provides mermaid diagram generation for any command that benefits from a visual representation. Picks the appropriate diagram type from the content signals, emits valid mermaid syntax, and self-skips when visualization would add no value. Activated by /fw:visualize directly, and on-demand by /fw:summarize, /fw:document, /fw:design, /fw:debug, /fw:implement, and /fw:reflect.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Multi-provider AI orchestration plugin — Let Claude orchestrate specialized agents (Codex, Gemini, Claude teammates) across structured workflows
Flywheel transforms Claude into an orchestrator that spawns specialized AI agents for different tasks. Instead of Claude doing everything alone, it delegates specific work to expert agents and synthesizes their outputs.
Think of it as: Claude is the project manager, sub-agents are the specialists.
flowchart LR
You["👤 You"] -->|"/fw:implement build auth"| Claude["🔵 Claude\nOrchestrator"]
subgraph TeamMode["🤝 Team Mode (parallel)"]
Researcher["🔵 researcher\nfw-researcher"]
Architect["🔵 architect\nfw-architect"]
end
subgraph SeqMode["⬇️ Sequential Mode (fallback)"]
Gemini["🟡 Gemini\nEcosystem Research"]
Codex["🔴 Codex\nArchitect + Builder"]
end
Claude -->|"AGENT_TEAMS=1"| TeamMode
Claude -->|"AGENT_TEAMS=0"| SeqMode
TeamMode -->|"findings"| Claude
SeqMode -->|"findings"| Claude
Claude -->|"synthesized result"| You
style Claude fill:#4A90E2,stroke:#2E5C8A,stroke-width:3px,color:#fff
style Codex fill:#E24A4A,stroke:#8A2E2E,stroke-width:2px,color:#fff
style Gemini fill:#E2B44A,stroke:#8A6E2E,stroke-width:2px,color:#fff
style Researcher fill:#4A90E2,stroke:#2E5C8A,stroke-width:2px,color:#fff
style Architect fill:#4A90E2,stroke:#2E5C8A,stroke-width:2px,color:#fff
# 1. Add this repository as a plugin marketplace
claude plugin marketplace add gmedali/flywheel-plugin
# 2. Install the plugin
claude plugin install fw
once you run claude, please run:
/fw:setup
/fw:setup will detect your providers, install personas, and optionally configure agent teams for parallel workflows. All /fw: commands are ready to use immediately.
Flywheel ships 10 native Claude Code sub-agents — persistent specialists that each /fw: command activates. They run in isolated context windows with curated tool access and model routing, replacing ad-hoc prompt strings with real behavioural mandates.
| Persona | Model | Used by | Role |
|---|---|---|---|
fw-architect | opus | /fw:design, /fw:implement | Commits to one approach, produces ADRs, file impact maps, risk registers |
fw-researcher | sonnet | Research phases | Filters external knowledge through your existing stack, rates every option |
fw-debugger | sonnet | /fw:debug | Forensic root-cause analysis — mandatory causal chain, no guesses |
fw-security-auditor | opus | /fw:harden | Offensive OWASP scanner — every finding has attack scenario + patch |
fw-code-reviewer | sonnet | /fw:review | Blocking vs advisory classification — every finding has file:line + fix |
fw-tdd-specialist | sonnet | /fw:tdd | Red→Green→Refactor enforcer — refuses implementation without failing test |
fw-test-generator | sonnet | /fw:test | P0→P3 risk-prioritised coverage — learns your conventions before writing |
fw-migration-engineer | sonnet | /fw:migrate | Batched migrations with mandatory rollback plans and validation gates |
fw-fe-designer | sonnet | /fw:fe-design | Distinctive, production-grade frontend interfaces — zero generic AI aesthetics |
fw-reflector | sonnet | /fw:reflect | Multi-dimensional quality synthesis — balances strengths with concerns, severity-sorted narrative reports |
Installation: /fw:setup runs scripts/install-personas.sh to copy all personas to ~/.claude/agents/. Upgrade anytime:
"${CLAUDE_PLUGIN_ROOT}/scripts/install-personas.sh" --force
Flywheel v1.1.0 adds native Claude-to-Claude parallel execution via the CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS feature flag. When enabled, complex commands spawn specialist teammates that work in parallel instead of sequentially.
Commands that gain parallel teammates:
| Command | Template | Parallel Phases | Teammates |
|---|---|---|---|
/fw:implement | research-and-plan | 1a+1b, 7+8 | researcher, architect, test-writer |
/fw:design | research-and-plan | 1+2 | researcher, architect |
/fw:review | parallel-review | All analysis (critical level) | security, quality, test reviewers |
/fw:reflect | parallel-analysis | Phase 2 aspects | 1 per aspect (up to 4) |
/fw:harden | parallel-scan | Phase 1 scans | owasp, dependency, secrets scanners |
/fw:migrate | parallel-migration-research | Phase 1b | framework-researcher, codebase-analyzer |
npx claudepluginhub gmedali/flywheel-plugin --plugin fwProduction-ready Claude Code configuration with role-based workflows (PM→Lead→Designer→Dev→QA), safety hooks, 44 commands, 19 skills, 8 agents, 43 rules, 30 hook scripts across 19 events, auto-learning pipeline, hook profiles, and multi-language coding standards
Full-stack orchestration with deployment, performance, security, and test automation
Mindful AI coding framework — discipline over cleverness. Skill + 21 slash commands + 8 specialist agents + 5 runtime hooks + 15 default checklists + Master Orchestrator + Gravity hub. Works on any model tier (Opus/Sonnet/Haiku). Integrates Claude Design for visual work.
An engineering team in a box for Claude Code. 12 specialized subagents (planner, fullstack-engineer, refactor-specialist, migration-engineer, frontend-designer, critic, vuln-verifier, debugger, db-expert, onboarder, tool-expert, web-researcher) plus 15 automation hooks (pre-commit secret scan, MCP health tracking, cost tracking, test runner, branch protection, large file warner, session summary, batch format, design quality, config protection, and more) wired by the P7/P9/P10 methodology with three red lines: closure discipline, fact-driven, exhaustiveness.
AI-powered development workflow automation - Phase-based planning, implementation orchestration, preflight code quality checks with security scanning, ship-it workflow, and development principles generator for CLAUDE.md
Multi-agent orchestration for code that matters.