By bryonjacob
FFFlow — a complete development lifecycle (plan, work, audit) in one Claude Code plugin. Scales from informal prose specs (L0) to full spec-driven development with mutation testing and RID traceability (L3). Same workflow at every level.
End-to-end onboarding for a new FFFlow project. Runs init-ffflow, stack-init, then hands off to plan-roadmap to phase out the work (greenfield or brownfield — roadmap figures out the right per-phase entry point).
Run the FFFlow keep-honest auditors against this project. Coordinator with per-type cartridges (CLAUDE.md freshness, spec drift, characterization tests, architecture conformance, complexity/refactor signals, and L3 RID traceability). Use --type to target one; --plan to convert findings into work.
Meta-skill that wraps any interactive skill for unattended execution. Reads project context to stand in for the user on routine decisions, with a decision log.
For an unspec'd codebase, walk it module-by-module to extract an evergreen spec (committed immediately) and produce a plan for characterization-test backfill.
GitHub Actions workflow templates for FFFlow projects — runs level-appropriate audits and quality gates as PR checks.
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.
Plan, work, audit — at the level of rigor your project actually needs.
FFFlow is a complete development lifecycle in a single Claude Code plugin. It scales from informal prose specs (L0) to full spec-driven development with mutation testing and RID traceability (L3). Same workflow at every level; what changes is the formality of the artifacts and strictness of the gates.
The plugin's invocation prefix is /fff: — short, distinctive, won't collide.
/plugin marketplace add /path/to/ffflow-plugin
/plugin install fff@ffflow
After install, every command is namespaced under /fff: (e.g., /fff:plan-chat, /fff:work-issue).
Three artifacts. Three roles. Three locations.
docs/specs/, specs/, or per-module CLAUDE.md./tmp/ffflow-plans/<project-hash>/<slug>/, never in the project tree.Bootstrap a new project (from project.md, README, or just an idea):
/fff:adopt-ffflow Detects greenfield vs. brownfield. Runs init + stack setup,
then hands off to /fff:plan-roadmap.
/fff:plan-roadmap Reads your vision; produces ordered phases with dependencies.
Per phase: design-slice (→ plan-chat), characterize, cleanup,
or recursive sub-roadmap.
Steady-state feature work (after bootstrap):
/fff:plan-chat Design one slice. Open decisions resolved inline before sealing.
/fff:plan-breakdown Cut the plan into 3–8 independently shippable chess moves.
/fff:plan-capture Write the chess moves to your issue tracker.
/fff:work-issue Execute one chess move end-to-end: red → green → spec → PR.
/fff:work-fanout Or run several in parallel across worktrees (one PR per issue).
/fff:work-epic Or run an epic's issues sequentially into one PR.
/fff:audit Periodically: check that specs, CLAUDE.md, tests still match reality.
Maintenance loop:
/fff:audit Find drift. Find gaps.
/fff:audit --plan Convert findings into a plan. Then breakdown → capture → work.
/fff:plan-roadmap (phase-level). Used for new projects, big features, or new directions./fff:plan-chat — design a slice./fff:characterize — extract specs from existing code./fff:audit --plan — convert drift into a plan.All four converge on plan-breakdown → plan-capture → work-*.
Single-slice changes skip tier 1; bug fixes go through plan-chat directly.
Pick the level your project needs. You can change levels later.
| Level | Spec format | Tests | Gates |
|---|---|---|---|
| L0 | Informal prose | Standard unit/integration | Lint + tests pass |
| L1 | Structured prose, hexagonal layering | + defect-driven spec entries | + CLAUDE.md audit |
| L2 | Gherkin 6 scenarios | + spec-first TDD | + spec audit, contract enforcement |
| L3 | Gherkin + RIDs + property-based | + mutation testing | + four quality gates, RID audit |
No config = L0. Declare your level in .ffflow/config.yaml to enable more.
Several skills carry their family of variants as on-demand cartridges, keeping the LLM's skill index uncluttered:
/fff:plan-capture — cartridges for github-issues (default), linear, jira, markdown./fff:stack — cartridges for python, typescript, java, rust, polyglot./fff:justfile — Tier 1–4 pattern cartridges (quality, security, advanced, polyglot)./fff:audit — cartridges for claude-md, spec, char-tests, architecture, refactor, tech-debt, rid.Configure once in .ffflow/config.yaml; the right cartridge loads when needed.
.ffflow/stack.yaml.Active development. See plugin/CLAUDE.md for the build phase status.
Apache License 2.0 — see LICENSE. Copyright 2026 Bryon Jacob.
npx claudepluginhub bryonjacob/ffflow --plugin fffJustfile standard interface management - maturity model, assessment, installation, and refactoring
Core development workflows and tooling standards. Stack configuration (Python/JS/Java), documentation (CLAUDE.md/MkDocs), workflow automation (justfile), CI/CD (GitHub Actions), and issue management.
AI-enhancement capabilities: session management, concise prompt writing (hemingwayesque), autonomous command execution (automate), and workflow orchestration.
Web development with Next.js 15+, React 19, static sites, Tailwind CSS v4 + CVA, and Formspree forms. Modern web UI patterns and deployment.
AI organization orchestrator — define agents, assign tasks, and run them as coordinated teams inside Claude Code
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.