agentleFS
JuanMarchettoJuanMarchetto/agent-skills/skills/founder-mode/SKILL.md

founder-mode

Skill5 starsChanged 8 months agoMITView on GitHub

End-to-end product builder with founder mindset. Orchestrates 9 phases: idea evaluation → architecture → TDD → implementation → E2E → verification → analysis → GTM. Produces a launched product with README, pitch, tweet thread, and Product Hunt draft. Use when: build product, founder mode, full lifecycle, idea to launch, end to end, build from scratch, MVP, product builder, launch kit, vibe coding with quality.

5 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Founder Mode — End-to-End Product Builder

From idea to launched product. Not just code — a product.

How It Works

/founder-mode "Build a CLI that converts markdown to slides"
        │
        ▼
  Phase 0: Bootstrap ─── config.json + state.json + mode detection
        │
        ▼
  Phase 1: Idea Evaluation ─── /architect dispatches 6 evaluators
        │
        ▼
  ══ GATE 1 ══ Go/No-Go decision ─── [continue|reject|iterate|edit]
        │
        ▼
  Phase 2: Architecture ─── system design, components, tech stack
        │
        ▼
  Phase 3: TDD Planning ─── decompose into test-first chunks
        │
        ▼
  Phase 4: Implementation ─── chunk-with-summary TDD loop
        │
        ▼
  ══ GATE 2 ══ Implementation review ─── [continue|iterate|stop|edit]
        │
        ▼
  Phase 5: E2E Testing ─── strategy by project type
        │
        ▼
  Phase 6: Verification ─── false positives + spec-to-test gap + arch drift
        │
        ▼
  ══ GATE 3 ══ Verification review ─── [continue|iterate|stop|edit]
        │
        ▼
  Phase 7: Post-Run Analysis ─── lessons + recommendations
        │
        ▼
  Phase 8: GTM Draft ─── README + pitch + tweets + PH + CHANGELOG
        │
        ▼
  DONE ─── all artifacts in .founder/

Phases

PhaseWhat It DoesAgent
0 - BootstrapConfig generation, mode detection, codebase scan (existing projects), lesson loadingbootstrapper.md
1 - EvaluationDispatches 6 parallel evaluators via /architect, extracts acceptance criteriaevaluator.md
2 - ArchitectureGenerates system design (new) or reverse-engineers from codebase (existing)architect-gen.md
3 - TDD PlanningDecomposes architecture into ordered test-first chunksplanner.md
4 - ImplementationTDD per chunk with cross-chunk summaries, supports re-entry for fixesimplementer.md
5 - E2E TestingGenerates and runs E2E tests matched to acceptance criteriae2e-runner.md
6 - VerificationFalse-positive detection, spec-to-test gap analysis, architecture drift checkverifier.md
7 - AnalysisPost-run pattern analysis, lesson extraction, recommendationsanalyzer.md
8 - GTM DraftREADME, pitch, tweet thread, Product Hunt listing, CHANGELOGgtm-drafter.md

Human Gates

Three checkpoints where you make Go/No-Go decisions:

  • Gate 1 (Post-Eval): Review the idea score. Continue, reject, or iterate with refined framing.
  • Gate 2 (Post-Implementation): Review test results and known issues. Continue or iterate to fix.
  • Gate 3 (Post-Verification): Review verification verdict. Continue to GTM or iterate to fix issues.

Each gate supports: [continue], [iterate], [stop], [edit]

Iteration starts with a cap of 3 (configurable). When reached, you're prompted to extend or move on.

Config

Generated at .founder/config.json during bootstrap. Key settings:

  • mode: "new" or "existing" (auto-detected)
  • maxIterations: iteration cap (default 3, extendable at gates)
  • gates: per-gate behavior — "pause", "auto", or "skip"
  • phases: enable/disable individual phases
  • tdd: strictness and coverage threshold
  • gtm: enable/disable individual GTM artifacts

Edit the config before saying go to customize the pipeline.

Modes

  • New Idea: Full pipeline from Phase 0 to Phase 8.
  • Existing Project: Pass a path (/founder-mode ./my-project). Skips eval, reverse-engineers architecture, plans around existing tests, extends code, respects existing test framework.

Dependencies

  • Required: architect skill (for Phase 1 evaluation)
  • Optional: learn-by-mistake (error memory), post-run-analysis (enhanced Phase 7)

Usage

/founder-mode "Build a task management API with natural language input"
/founder-mode "./my-existing-project"
/founder-mode resume

Crash Recovery

State is persisted to .founder/state.json after every phase and gate decision. If your session crashes, run /founder-mode resume to pick up exactly where you left off — including mid-gate crashes.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.