agent-flywheel

Plugin agent-flywheel from marketplace agent-flywheel in burningportra/agent-flywheel-plugin. Commands read from the repo's manifest — 23 days ago.

Skills (47)

• _template

  One-sentence summary of what this skill does. Start with a verb, keep it specific enough that Claude can decide when to invoke it.

• agent-flywheel

  Use when starting, resuming, or operating the Agent Flywheel multi-agent workflow (scan, discover, plan, beads, NTM swarm implementation, review, and wrap-up), running Poteto (pstack) mode (/why, /interrogate, /architect, SHA-coupled verification ledger), or syncing and troubleshooting the Antigravi

• architect

  Sketch types, signatures, and module structure before code, then stay in the loop while implementation fills in. Use for /architect, 'architect this', 'design this', or non-trivial work where jumping to code would lock in the wrong shape.

• brainstorming

  You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

• caam

  Manage AI coding CLI accounts with sub-100ms switching. Use when hitting rate limits on Claude Max, GPT Pro, or Gemini Ultra subscriptions and need instant account swapping without browser OAuth.

• cass-memory

  CASS Memory System (cm) for procedural memory. Use when starting non-trivial tasks, learning from past sessions, building playbooks, or preventing repeated mistakes via trauma guard.

• cass

  Mine past agent sessions for working prompts, decisions, and patterns. Use when "what did I ask?", "find that prompt", session archaeology, or agent history.

• changelog-md-workmanship

  Rebuild CHANGELOG.md files and release histories from git, tags, releases, and issue trackers. Use when writing changelogs, version timelines, or agent-facing project history summaries.

• docs-de-slopify

  Remove telltale signs of AI-generated "slop" writing from documentation. Use when polishing README files, API docs, or any public-facing text to sound authentically human.

• dsr

  Doodlestein Self-Releaser - fallback release infrastructure for when GitHub Actions is throttled. Local builds, cross-platform releases, supply chain security. Use when: GH Actions slow, local release, build hosts, dsr command.

• flywheel-audit

  Audit the codebase for bugs, security issues, test gaps, and dead code.

• flywheel-cleanup

  Clean up orphaned git worktrees from crashed or stopped sessions.

• flywheel-compound-refresh

  Sweep docs/solutions/ to consolidate, replace, or archive stale and duplicate learnings — never auto-deletes.

• flywheel-doctor

  One-shot diagnostic of every flywheel dependency — MCP connectivity, Agent Mail liveness, br/bv/ntm/cm binaries, node version, git status, dist-drift, orphaned worktrees, and checkpoint validity. Use when debugging toolchain issues, before starting a new session, after /flywheel-cleanup, or as a CI

• flywheel-drift-check

  Check if the codebase has drifted from the implementation plan.

• flywheel-duel

  Run a state-aware /dueling-idea-wizards duel inside the flywheel. Routes artifacts into discovery, plan, or review pipelines based on the current phase. Use when the user invokes /flywheel-duel directly, or when start.md routes here from a Discover/Plan/Review menu.

• flywheel-fix

  Fast path to apply a targeted fix without running the full flywheel.

• flywheel-healthcheck

  Full health check of the codebase and flywheel dependencies.

• flywheel-mine-patterns

  Mine cass session history for recurring workflows and propose them as new skills, refinements to existing skills, or flywheel pipeline tweaks. Use when "what should we automate next?", "find repeated rituals", or after a long stretch of agent work to harvest tacit knowledge into beads.

• flywheel-reality-check

  Strategic gap analysis between vision (AGENTS.md / README.md / plan docs) and what's actually implemented. Converts gaps into beads, optionally launches a swarm. Use when "reality check", "where are we really", "gap analysis", "did we drift", or before declaring a long-running project done.

• flywheel-refine-skill

  Refine a specific agent skill using session evidence and feedback.

• flywheel-refine-skills

  Review and improve all loaded agent skills based on session patterns and feedback.

• flywheel-research

  Deep research on an external GitHub repository to extract implementation insights.

• flywheel-rollback

  Roll back a completed bead implementation.

• flywheel-scan

  Targeted scan of specific paths or concerns without a full flywheel scan.

• flywheel-setup

  Set up flywheel prerequisites for this project.

• flywheel-speedup

  Make a slow endpoint, flow, job, or service fast. Baselines and profiles it, freezes behavior with real-data e2e tests, then either optimizes in place or, when the slowness is structural, plans a same-API re-architecture, swarms it through the flywheel, gauntlets it under load, and cuts over behind

• flywheel-status

  Show current flywheel status, bead progress, and inbox messages.

• flywheel-stop

  Stop the current flywheel session and reset state.

• flywheel-swarm-status

  Check the status of running swarm agents and bead progress.

• flywheel-swarm-stop

  Stop all running swarm agents and release their file reservations.

• flywheel-swarm

  Launch a parallel swarm of agents to implement multiple beads simultaneously.

• flywheel-tool-feedback

  Submit structured feedback about a tool or command to improve future flywheel sessions.

• frontend-design

  Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications. Generates creative, polished code that avoids generic AI aesthetics.

• gdb-for-debugging

  Debug running processes with GDB, strace, and /proc inspection. Use when: process hang, spin loop, segfault, crash, backtrace, attach debugger, ptrace, core dump, thread starvation, accept loop, deadlock, busy wait, or diagnose why a process is stuck.

• grill-with-docs

  Relentless goal/design interview that sharpens framing and writes durable docs (brainstorm artifact, ADRs, glossary) as decisions crystallize. Use when refining a flywheel goal, pressure-testing scope, or "grill with docs".

• idea-wizard

  Generate and operationalize improvement ideas for projects. Use when brainstorming features, planning improvements, creating beads from ideas, or "what should we build next".

• interrogate

  Use for \"interrogate\", \"adversarial review\", \"multi-model review\", \"challenge this\", \"stress test this code\", \"find blind spots\", or \"tear this apart\". Multiple LLM reviewers challenge changes from independent angles.

• memory

  Search, store, or manage CASS long-term memory.

• multi-model-triangulation

  Cross-validate decisions using multiple AI models (Codex, Gemini, Grok). Use when "get a second opinion", evaluating approaches, or high-stakes decisions.

• poteto-mode

  poteto's agent style for concise, detailed responses, deliberate subagents, unslopped prose, simple code, and verified work. Use for poteto, /poteto-mode, or requests to work in this style.

• slb

  Simultaneous Launch Button - Two-person rule for destructive commands. Use when coordinating dangerous operations between agents, requiring peer review for rm -rf, git push --force, kubectl delete, DROP TABLE, or terraform destroy.

• start

  Start or resume the full agentic coding flywheel. Drives the complete workflow: scan → discover → plan → implement → review.

• ubs-workflow

  Comprehensive code review workflow using Ultimate Bug Scanner (UBS). Use when reviewing code, scanning for bugs, validating AI-generated code, or running pre-commit quality checks. - review code - scan for bugs - code review - ubs - find bugs - security audit - lang-typescript - lang-python - lang-r

• ui-polish

  Iterative UI/UX polishing for Stripe-level visual quality. Use when app already works and looks decent, wanting to elevate to world-class through multiple passes. Not for complete overhauls.

• ui-ux-polish

  Run a structured UI/UX polish pass on implemented code. Spawns a scrutiny agent to generate 15-30 improvement suggestions, presents them numbered for user selection, converts selected ones to beads, and triggers implementation. Use after initial implementation is complete.

• why

  Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Discovers available MCPs and queries each evidence category (source control, issue tracker, long-form docs, real-time chat, infrastructure observability, error tracking, prod

Slash commands (26)

• /agent-flywheel:flywheel-audit

  Audit the codebase for bugs, security issues, test gaps, and dead code.

• /agent-flywheel:flywheel-bead-viewer

  Open the bead-graph visualizer in your browser.

• /agent-flywheel:flywheel-cleanup

  Clean up orphaned git worktrees from crashed or stopped sessions.

• /agent-flywheel:flywheel-compound-refresh

  Sweep docs/solutions/ for stale, duplicate, and contradictory learnings — Keep / Update / Consolidate / Replace / Delete.

• /agent-flywheel:flywheel-doctor

  One-shot diagnostic of every flywheel dependency. Reports MCP connectivity, Agent Mail, br/bv/ntm/cm, node, git, dist drift, orphaned worktrees, and checkpoint validity with a single glyph per line.

• /agent-flywheel:flywheel-drift-check

  Check if the codebase has drifted from the implementation plan.

• /agent-flywheel:flywheel-duel

  Run a state-aware /dueling-idea-wizards duel. Routes artifacts into the flywheel pipeline (discovery, plan, review) based on current phase.

• /agent-flywheel:flywheel-fix

  Fast path to apply a targeted fix without running the full flywheel.

• /agent-flywheel:flywheel-healthcheck

  Full health check of the codebase and flywheel dependencies.

• /agent-flywheel:flywheel-reality-check

  Strategic gap analysis between vision (AGENTS.md / README.md / plan docs) and actual implementation. Reads docs, investigates code, applies /reality-check-for-project exhaustively, converts gaps to tagged beads, optionally launches a 3+3 swarm.

• /agent-flywheel:flywheel-refine-skill

  Refine a specific agent skill using session evidence and feedback.

• /agent-flywheel:flywheel-refine-skills

  Review and improve all loaded agent skills based on session patterns and feedback.

• /agent-flywheel:flywheel-research

  Deep research on an external GitHub repository to extract implementation insights.

• /agent-flywheel:flywheel-rollback

  Roll back a completed bead implementation.

• /agent-flywheel:flywheel-scan

  Targeted scan of specific paths or concerns without a full flywheel scan.

• /agent-flywheel:flywheel-setup

  Set up flywheel prerequisites for this project.

• /agent-flywheel:flywheel-speedup

  Make a slow endpoint, flow, job, or service fast. Baselines and profiles it, freezes behavior with real-data e2e tests, then optimizes in place or plans a same-API re-architecture, swarms it, gauntlets it under load, and cuts over behind a canary with automatic rollback.

• /agent-flywheel:flywheel-status

  Show current flywheel status, bead progress, and inbox messages.

• /agent-flywheel:flywheel-stop

  Stop the current flywheel session and reset state.

• /agent-flywheel:flywheel-swarm-status

  Check the status of running swarm agents and bead progress.

• /agent-flywheel:flywheel-swarm-stop

  Stop all running swarm agents and release their file reservations.

• /agent-flywheel:flywheel-swarm

  Launch a parallel swarm of agents to implement multiple beads simultaneously.

• /agent-flywheel:flywheel-tool-feedback

  Submit structured feedback about a tool or command to improve future flywheel sessions.

• /agent-flywheel:grill-with-docs

  Relentless goal/design interview that writes brainstorm + optional ADRs/glossary, then hands off to flywheel planning.

• /agent-flywheel:memory

  Search, store, or manage CASS long-term memory.

• /agent-flywheel:start

  Start or resume the full agentic coding flywheel. Drives the complete workflow: scan → discover → plan → implement → review.

MCP servers (2)

• agent-mail

  connects to: http://127.0.0.1:8765/mcp

• agent-flywheel

  runs: node ${CLAUDE_PLUGIN_ROOT}/mcp-server/scripts/launch-server.mjs

Hooks (7)

• SessionStart

  bash "${CLAUDE_PLUGIN_ROOT}/hooks/_run-hook.sh" node "${CLAUDE_PLUGIN_ROOT}/hooks/startup.js"

• PreToolUse

  node "${CLAUDE_PLUGIN_ROOT}/hooks/agent-mail-guard.js"

• PreToolUse

  node "${CLAUDE_PLUGIN_ROOT}/hooks/agent-mail-guard.js"

• PreToolUse

  node "${CLAUDE_PLUGIN_ROOT}/hooks/agent-mail-guard.js"

• Stop

  bash "${CLAUDE_PLUGIN_ROOT}/hooks/_run-hook.sh" node "${CLAUDE_PLUGIN_ROOT}/hooks/agent-mail-guard.js" --release-reservations

• SubagentStop

  bash "${CLAUDE_PLUGIN_ROOT}/hooks/_run-hook.sh" node "${CLAUDE_PLUGIN_ROOT}/hooks/agent-mail-guard.js" --release-reservations

• PostToolUse

  echo 'Bead approval processed. Run /agent-flywheel:flywheel-status to check progress.'