convez.ai
SkillProdutividadeIntermediário
SanketSapkal

Founder Skills

SanketSapkal/founder-skills

Interrogue ideias antes de construir: BUILD, KILL ou PIVOT

Ver no GitHub
InicialPluginScore · de 0 a 100

Pipeline de interrogação adversarial de ideias de startup que produz um veredito explícito BUILD/KILL/PIVOT. Diferencia-se por questionar primeiro e validar depois, evitando vieses de confirmação. Integra-se com Claude Code, Codex, Cursor e outros editores de IA.

Instalação

$npx convez-ai SanketSapkal/founder-skills

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
14
Forks
3
Acompanhando
1
Saúde
28%
Último commit
há 6 meses
Versão
—
Issues abertas
0
Contribuidores
2

Casos de uso

  • Validar ideias de startup com pensamento crítico antes de investir tempo
  • Descobrir riscos ocultos e assumptions não testadas em conceitos de negócio
  • Gerar roadmap estruturado (canvas, unit economics, GTM) apenas para ideias viáveis

Veredito

Ferramenta robusta e bem arquitetada para founders que querem feedback adversarial genuíno antes de comprometer recursos. A pipeline sequencial e o aviso sobre vieses de LLM mostram maturidade de design. Recomendado para quem quer rigor estratégico, não validação fácil.

README

A tradução para o português deste README ainda está na fila.

founder-skills

Adversarial idea interrogation followed by structured business modeling — in one chained pipeline.

Every existing startup skills repo is generative: you bring an idea and it helps you fill it out. This repo starts by trying to kill the idea first, then builds out the survivors.

Existing repos:   Interrogate OR Model
This repo:        Interrogate THEN Model

The Key Differentiator

/verdict — a synthesis skill that reads all prior outputs and returns an explicit BUILD / KILL / PIVOT judgment with structured reasoning. No other skills repo has this.

Pipeline

[/problem-discovery] ──> [/problem-statement] ──> /pressure-test ──> /customer-archetype ──┬──> /founder-fit
   (optional, if no                                    ↑                                    ├──> /scope-mode ──> /startup-canvas ──> /unit-economics
    specific idea yet)                          ⚠ run in fresh                              │                          │
                                                 conversation                               ├──> /market-size ─────────┤
                                                 to avoid bias                              │                          ├──> /regulatory-risk (regulated industries)
                                                                                            │                          ├──> /battle-cards
                                                                                            │                          └──> /assumption-map ──> /pretotype
                                                                                            └──────────────────────────────────────────────────────┘
                                                                                                                                                  ↓
All outputs ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────> /verdict
/verdict (BUILD) ──┬───────────────────────────────> /build-brief → coding agent
                   ├───────────────────────────────> /pitch-deck, /investor-lens, /elevator-pitch, /objection-map
                   └───────────────────────────────> /gtm-plan, /oneliner, /cold-outreach, /content-strategy
/verdict (KILL/PIVOT) ─────────────────────────────> /pivot-paths

⚠ Bias warning: /problem-discovery and /problem-statement help shape the problem — they are not adversarial. After running them, switch to a fresh Claude Code conversation before running /pressure-test. Otherwise the agent that just helped you frame the problem will be biased toward validating it, undercutting the adversarial review. This is a fundamental limit of LLM self-review, not a Claude-specific issue.

Every skill writes a structured output file to ./founder-outputs/. Downstream skills read those files. /verdict reads all of them.

Quick Start

  1. Install the plugin (see below)
  2. Open a project in Claude Code
  3. Say "I have an idea" or run /pressure-test
  4. Follow the pipeline — each skill recommends the next step

Installation

As a Claude Code plugin (recommended)

claude plugin add /path/to/founder-skills

Multi-platform install

The setup script auto-detects installed coding agents and installs the appropriate format for each:

git clone https://github.com/YOUR_USERNAME/founder-skills.git
cd founder-skills
./setup --auto     # detect and install for all found agents

Or install for specific platforms:

./setup --claude global   # Claude Code — symlink to ~/.claude/skills/
./setup --claude local    # Claude Code — symlink to ./.claude/skills/
./setup --codex           # Codex CLI — generate AGENTS.md
./setup --cursor          # Cursor — generate .cursorrules
./setup --kiro            # Kiro — copy SKILL.md to ~/.kiro/skills/
./setup --opencode        # OpenCode — copy SKILL.md to .opencode/skills/
./setup --windsurf        # Windsurf — generate .windsurf/rules/
./setup --all             # install for every supported platform

Supported Platforms

PlatformFormat GeneratedInstall Location
Claude CodeSKILL.md (native)~/.claude/skills/ or .claude/skills/
Codex CLIAGENTS.md (plain markdown)project root
Cursor.cursorrules (markdown)project root
KiroSKILL.md (copied)~/.kiro/skills/
OpenCodeSKILL.md (copied).opencode/skills/
Windsurf.windsurf/rules/*.mdproject root

Skills Reference

SkillPurposeReadsWrites
/problem-discoverySurface candidate problems when no specific idea exists yet—problem-discovery-output.md
/problem-statementPolish a vague problem into a precise JTBD statementproblem-discovery (optional)problem-statement-output.md
/pressure-testAdversarial interrogation — kill bad ideas earlyproblem-statement (optional)pressure-test-output.md
/customer-archetypeForce precise customer definition + buying committee mappingpressure-testcustomer-archetype-output.md
/founder-fitAssess founder-market fit and unfair advantagepressure-test, archetypefounder-fit-output.md
/scope-modeCalibrate ambition: EXPAND / SELECTIVE / HOLD / REDUCEpressure-test, archetypescope-output.md
/market-sizeBottom-up TAM / SAM / SOMpressure-test, archetypemarket-size-output.md
/startup-canvasBusiness model + value propositionpressure-test, archetype, scopestartup-canvas-output.md
/unit-economicsCAC / LTV / payback / gross margin — viability mathcanvas, archetype, market-size, gtm-planunit-economics-output.md
/regulatory-riskCompliance exposure and licensing requirementscanvas, archetype, market-sizeregulatory-risk-output.md
/battle-cardsCompetitive intelligence + switching costs + positioningarchetype, canvasbattle-cards-output.md
/assumption-mapSurface and prioritize riskiest assumptionscanvas, market-sizeassumption-map-output.md
/pretotypeDesign validation experimentsassumption-mappretotype-output.md
/verdictBUILD / KILL / PIVOT synthesisALL outputsverdict-output.md
/pivot-paths3 concrete pivot directionsverdict + all outputspivot-paths-output.md
/10x-why-nowThiel's "why now" formalizedpressure-test, battle-cardswhy-now-output.md
/analogous-cosStructurally similar companies in other marketspressure-test, canvasanalogous-cos-output.md
/build-briefStructured handoff for coding agentsALL outputs (BUILD verdict)build-brief-output.md
Investment Pitch
/pitch-deckSequoia narrative arc from all outputsALL outputspitch-deck-output.md
/investor-lensScore through seed angel, Series A, corporate VC lensesverdict, canvas, market-sizeinvestor-lens-output.md
/elevator-pitch30-sec, 2-min, 10-min pitch versionscanvas, archetype, battle-cardselevator-pitch-output.md
/objection-map8-10 hardest investor questions with answersverdict, assumption-mapobjection-map-output.md
Sales & Marketing
/gtm-planBullseye framework — rank 19 channels, test top 3canvas, archetypegtm-plan-output.md
/onelinerOne-sentence value prop with scoring criteriacanvas, archetype, battle-cardsoneliner-output.md
/cold-outreachPersonalized 3-email sequences from archetypearchetype, canvascold-outreach-output.md
/content-strategy30-day content plan matched to archetype sourcesarchetype, canvas, battle-cardscontent-strategy-output.md

CLAUDE.md Snippet

Add this to your project's CLAUDE.md to give Claude pipeline awareness:

## Founder Skills Pipeline

This project uses the founder-skills pipeline for idea validation.

### Output Directory
All skill outputs are saved to `./founder-outputs/`. Each skill reads
prior outputs from this directory to build on earlier analysis.

### Pipeline Sequence
0a. `/problem-discovery` — (optional) Surface candidate problems if no idea yet
0b. `/problem-statement` — (optional) Polish a vague problem before pressure-testing
1.  `/pressure-test` — Adversarial interrogation (run first if you have an idea)
2.  `/customer-archetype` — Define the specific customer
3.  `/founder-fit` — Assess founder-market fit
4.  `/scope-mode` — Calibrate ambition level
5.  `/market-size` — Bottom-up TAM/SAM/SOM
6.  `/startup-canvas` — Business model + value proposition
7.  `/unit-economics` — CAC / LTV / payback / gross margin
8.  `/regulatory-risk` — Compliance exposure (regulated industries)
9.  `/battle-cards` — Competitive intelligence
10. `/assumption-map` — Risk prioritization
11. `/pretotype` — Validation experiments
12. `/verdict` — BUILD / KILL / PIVOT judgment

> **Bias warning:** if you run `/problem-discovery` or `/problem-statement`, switch to a fresh Claude Code conversation before `/pressure-test` — otherwise the agent that helped frame the problem will be biased toward validating it.

Books Referenced

BookAuthorUsed By
The Mom TestRob Fitzpatrick/pressure-test, /customer-archetype
Zero to OnePeter Thiel/pressure-test, /scope-mode, /10x-why-now
Competing Against LuckClayton Christensen/pressure-test, /customer-archetype, /startup-canvas
Lean AnalyticsCroll & Yoskovitz/market-size
The Lean StartupEric Ries/startup-canvas, /pretotype, /scope-mode
Business Model GenerationOsterwalder/startup-canvas
$100M OffersAlex Hormozi/startup-canvas
Continuous Discovery HabitsTeresa Torres/assumption-map
Obviously AwesomeApril Dunford/battle-cards
Crossing the ChasmGeoffrey Moore/battle-cards, /10x-why-now
Pretotype ItAlberto Savoia/pretotype
The Hard Thing About Hard ThingsBen Horowitz/verdict
Shape UpBasecamp/build-brief
InspiredMarty Cagan/build-brief
Venture DealsFeld & Mendelson/investor-lens, /pitch-deck
The Fundraising RulesMark MacLeod/objection-map
TractionWeinberg & Mares/gtm-plan
$100M LeadsAlex Hormozi/gtm-plan, /cold-outreach, /content-strategy
Predictable RevenueRoss & Tyler/cold-outreach
Secrets of Sand Hill RoadScott Kupor/verdict
Ver o original no GitHub

Bom para

Relacionados