Startup Design
Multi-phase startup validation: research, lean canvas, financials, honest go/no-go scorecard
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 21, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 2dc2d3f
Fetched SKILL.md (31.8KB) plus references/honesty-protocol.md and frameworks.md from raw GitHub; spot-checked 3 referenced files (output-guidelines, frameworks, research-wave-1) all HTTP 200, a bogus path 404 as control. Ran the install into a temp HOME: clone+copy placed SKILL.md at ~/.claude/skills/startup-design/SKILL.md with all 15 reference files. For output I produced a real Lean Canvas for a freelancer invoice-reminder app twice — a baseline "make a lean canvas" version and a skill-faithful version following frameworks.md fill order plus the honesty protocol. Concrete deltas: baseline listed 1 problem and cheerled ("promising, build an MVP"); skill version listed 3 problems with existing alternatives, used the "We help [customer]…by [mechanism]" UVP formula, AARRR key metrics, [Data]/[Assumption]/[Opinion] labels, and Red/Yellow Flags that named the feature-not-a-company risk, refused to invent an unfair advantage, and recommended 5 interviews before building.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 7/10
- Docs & honesty 4/5
What Startup Design does
An 8-phase startup validation workflow that takes a raw idea through market research, competitive analysis, customer discovery, strategy, brand, product, financial projections, and validation experiments, writing ~30 structured markdown deliverables with a PROGRESS.md checkpoint so long runs survive interruption. Triggers when a user has a startup idea to explore, wants to validate a business concept, or asks for a lean canvas, market sizing, competitive landscape, positioning, or go-to-market plan. A radical-honesty protocol labels claims [Data]/[Estimate]/[Assumption]/[Opinion] and forces Red/Yellow flags plus a clear verdict rather than cheerleading.
How to install Startup Design
git clone --depth 1 https://github.com/ferdinandobons/startup-skill.git /tmp/startup-design-src
mkdir -p ~/.claude/skills
cp -R /tmp/startup-design-src/startup-design ~/.claude/skills/startup-design
# SKILL.md lives in the startup-design/ subdir of a 4-skill repo; the copy above installs just this one.
# Plugin alternative (installs all 4 startup skills):
# claude plugin marketplace add ferdinandobons/startup-skill
# claude plugin install startup@startup-skill
# Runtime: heavy skill. Full mode uses WebSearch + parallel Agent subagents across 4 research waves.
# Falls back to a Knowledge-Based mode (reduced confidence, flagged) if WebSearch is unavailable.
# Phase 3.7 (customer interviews) needs real calendar time; can be deferred. No API key required.
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Startup Design
-
/startup-designMulti-phase startup validation: research, lean canvas, financials, honest go/no-go scorecard
It also activates on plain-language prompts like these:
-
Validate my startup idea with market sizing -
Build a lean canvas for my new business -
Help me define brand positioning and go-to-market
Frequently asked questions
- Is the Startup Design skill free?
- Yes. The skill itself is free from ferdinandobons/startup-skill. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Startup Design work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Tested · Works. Fetched SKILL.md (31.8KB) plus references/honesty-protocol.md and frameworks.md from raw GitHub; spot-checked 3 referenced files (output-guidelines, frameworks, research-wave-1) all HTTP 200, a bogus path 404 as control. Ran the install into a temp HOME: clone+copy placed SKILL.md at ~/.claude/skills/startup-design/SKILL.md with all 15 reference files. For output I produced a real Lean Canvas for a freelancer invoice-reminder app twice — a baseline "make a lean canvas" version and a skill-faithful version following frameworks.md fill order plus the honesty protocol. Concrete deltas: baseline listed 1 problem and cheerled ("promising, build an MVP"); skill version listed 3 problems with existing alternatives, used the "We help [customer]…by [mechanism]" UVP formula, AARRR key metrics, [Data]/[Assumption]/[Opinion] labels, and Red/Yellow Flags that named the feature-not-a-company risk, refused to invent an unfair advantage, and recommended 5 interviews before building.
- What is the Startup Design SkillProof Score?
- 8.4/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 4/5.
- How do I install Startup Design?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Startup Design with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.