{
  "markdown": "# Testing Business Ideas: an Agent Skill\n\nAn [Agent Skill](https://agentskills.io) that turns your coding agent into a startup\nvalidation coach. It follows the open SKILL.md standard, so it works natively in\n**Claude Code, OpenAI Codex, Gemini CLI**, and other compatible agents (Cursor,\nGitHub Copilot, and more). Give it a business idea and it walks you through the\nmethod from *Testing Business Ideas* by David Bland and Alex Osterwalder\n(Strategyzer, 2020): break the idea into testable hypotheses, sort them by risk,\nfind the ones that would kill the idea if they're wrong, and pick concrete\nexperiments to run this week instead of building on faith.\n\nIt ships with the complete 44-experiment library, the assumptions-mapping and\nhypothesis framework, a plain-English glossary, and fill-in Test Card / Learning Card\ntemplates.\n\n> Not affiliated with or endorsed by the authors or Strategyzer. This is an\n> independent educational tool that summarizes and restructures the method for\n> coaching use. If you find it useful, buy the book. It is very good.\n\n## What it does\n\nYou bring an idea. The skill runs six steps:\n\n1. **Surface assumptions** hiding inside your idea.\n2. **Write testable hypotheses** (\"We believe that...\", precise and discrete).\n3. **Map and prioritize** them by importance and evidence, so you test the dangerous\n   unknowns first, not the comfortable ones.\n4. **Pick experiments** from the library, cheapest and fastest first, matched to your\n   business type.\n5. **Design the test** with a Test Card and a pass/fail number set in advance.\n6. **Read the results** into a persevere / pivot / kill decision.\n\nThe throughline: evidence from what people *do* beats what they *say*, and you should\nreduce uncertainty before you build.\n\n## Example\n\n```\nYou:    I want to build a subscription box of science kits for parents.\n\nClaude: Here's the idea as a rough business model, and the assumptions it's\n        resting on... The riskiest one, important and with zero evidence, is\n        that parents will pay monthly rather than buy one-off. Let's test that\n        first, before anything gets built.\n\n        Hypothesis: \"Millennial parents with kids 5-9 will pay $15/month for\n        curated science kits matched to their kid's grade level.\"\n\n        Cheapest strong test: a landing page with a real Subscribe button that\n        takes a card, driven by ~$200 of ads. Evidence strength 4/5 because\n        it's real payment intent, not a survey answer. Here's your Test Card...\n```\n\n## Install\n\nPick your agent. The same skill folder serves all of them.\n\n### Quick install, any agent (skills.sh)\n\n[skills.sh](https://skills.sh) (by Vercel Labs) installs skills into 70+ agents,\nincluding Claude Code, Codex, Gemini CLI, and Cursor:\n\n```bash\nnpx skills add nimitbhargava/testing-business-ideas-skill\n```\n\n### OpenAI Codex\n\nCodex supports Agent Skills natively. Inside Codex, run the installer skill:\n\n```\n$skill-installer install https://github.com/nimitbhargava/testing-business-ideas-skill/tree/main/skills/testing-business-ideas\n```\n\nOr copy the folder manually to `~/.agents/skills/` (or `~/.codex/skills/` on older\nbuilds). Invoke with `$testing-business-ideas` or just describe your idea.\n\nThe repo is also a Codex plugin marketplace (the `.codex-plugin/` and `.agents/`\nmanifests):\n\n```\n/plugin marketplace add nimitbhargava/testing-business-ideas-skill\n/plugin install testing-business-ideas@nimit-skills\n```\n\n### Gemini CLI\n\nGemini CLI supports Agent Skills natively:\n\n```bash\ngemini skills install https://github.com/nimitbhargava/testing-business-ideas-skill.git --consent\n```\n\nThe repo is also a Gemini CLI extension (this is what the root\n`gemini-extension.json` is for):\n\n```bash\ngemini extensions install https://github.com/nimitbhargava/testing-business-ideas-skill --auto-update\n```\n\n### Claude Code\n\nTwo ways: install it as a plugin from the marketplace (one command, auto-updates), or\ncopy the skill folder in by hand. The repo doubles as its own Claude Code plugin\nmarketplace.\n\n#### Option A: plugin marketplace (recommended)\n\nInside Claude Code, run:\n\n```\n/plugin marketplace add nimitbhargava/testing-business-ideas-skill\n/plugin install testing-business-ideas@nimit-skills\n```\n\nThat is it. The skill loads immediately and auto-fires when you describe a business\nidea. In the `/` menu it appears under the plugin namespace (Claude Code namespaces\nplugin skills, so it shows as `/testing-business-ideas:testing-business-ideas`); you\nrarely need to type it, just describe your idea.\n\nUpdate or remove it later with:\n\n```\n/plugin marketplace update nimit-skills\n/plugin uninstall testing-business-ideas@nimit-skills\n```\n\n#### Option B: copy the skill folder\n\nThis gives you the clean bare command `/testing-business-ideas` (no plugin namespace).\n\n**Personal (available in every project):**\n\n```bash\ngit clone https://github.com/nimitbhargava/testing-business-ideas-skill.git\ncp -r testing-business-ideas-skill/skills/testing-business-ideas ~/.claude/skills/\n```\n\n**Project-scoped (available only in one repo, shareable with your team):**\n\n```bash\ngit clone https://github.com/nimitbhargava/testing-business-ideas-skill.git\ncp -r testing-business-ideas-skill/skills/testing-business-ideas /path/to/your/project/.claude/skills/\n```\n\n**Or use the installer** (defaults to a personal install):\n\n```bash\ngit clone https://github.com/nimitbhargava/testing-business-ideas-skill.git\ncd testing-business-ideas-skill\n./install.sh            # installs to ~/.claude/skills/\n./install.sh /path/to/project   # installs to that project's .claude/skills/\n```\n\nRestart Claude Code (or start a new session) after a manual install so it picks up the\nskill. With Option B the folder name becomes a slash command automatically: type\n`/testing-business-ideas`, optionally with your idea inline\n(`/testing-business-ideas a subscription box of science kits for parents`).\n\n### Claude apps (claude.ai and Desktop)\n\nThere is no slash-command syntax on claude.ai or the Desktop app, and `~/.claude/skills/`\non your machine is Claude Code only. To use this skill there, add it as a skill through\nthe Customize menu, then it auto-invokes when you describe an idea. For cloud sessions\n(routines and similar), commit the `skills/testing-business-ideas/` folder into your\nrepo's `.claude/skills/` so the session loads it.\n\n### API and your own agents\n\nThe same folder is a valid [Agent\nSkill](https://docs.claude.com/en/docs/agents-and-tools/agent-skills/overview). Upload\nit as a skill in the Claude Developer Platform, or load `SKILL.md` as a system prompt\nwith the reference files attached if you are wiring it into your own agent. The content\nis plain Markdown and JSON, no dependencies, no build step.\n\n## Usage\n\nTwo ways to trigger it in Claude Code:\n\n- **Slash command:** `/testing-business-ideas` (Option B install) or the namespaced\n  `/testing-business-ideas:testing-business-ideas` (plugin install), optionally with\n  your idea after it.\n- **Just talk:** describe an idea and Claude fires the skill on its own. For example:\n  - \"Help me validate this business idea: ...\"\n  - \"What's the riskiest assumption in this and how do I test it?\"\n  - \"What experiment should I run to find out if people will pay for this?\"\n  - \"I ran the test and got these results, should I persevere or pivot?\"\n\nYou do not need to have read the book. The skill carries the method.\n\n## What's inside\n\n```\ntesting-business-ideas-skill/\n├── .claude-plugin/\n│   ├── marketplace.json              Claude Code plugin marketplace\n│   └── plugin.json                   Claude Code plugin manifest\n├── .codex-plugin/\n│   └── plugin.json                   OpenAI Codex plugin manifest\n├── .agents/\n│   └── plugins/marketplace.json      Codex plugin marketplace\n├── gemini-extension.json             Gemini CLI extension manifest\n├── README.md\n├── LICENSE\n├── install.sh\n└── skills/\n    └── testing-business-ideas/       <- the skill (Option B copies this into ~/.claude/skills/)\n        ├── SKILL.md                  the coaching workflow Claude follows\n        ├── reference/\n        │   ├── method.md             the framework: risks, hypotheses, evidence, decide\n        │   ├── experiment-library.md all 44 experiments, scannable, with selection rules\n        │   ├── glossary.md           plain-English definitions of the vocabulary\n        │   ├── mindset.md            the 8 pitfalls, leadership, funding stages\n        │   └── testing-business-ideas.json  structured source data\n        └── templates/\n            ├── test-card.md          design a test before running it\n            └── learning-card.md      turn results into a decision\n```\n\nThe skill uses progressive disclosure: `SKILL.md` stays short, and Claude loads a\nreference file only when a step needs it. That keeps the skill cheap to run and easy\nto read.\n\n## Coverage and provenance\n\nThe skill covers the whole book:\n\n- **All 44 experiments** (pp.106-306), with cost, setup time, run time, evidence\n  strength, tested risks, and type **transcribed from the book's own rating tables,\n  profile pages, and selection charts** (vector-extracted from the page data and\n  cross-checked visually), using the book's printed scale calibration. The few places\n  the book contradicts itself are adjudicated by 2-of-3 majority across those sources\n  and listed in the JSON's `library_notes`.\n- **All 10 of the book's named case studies** (Product Hunt, Intuit, Palm, NSA\n  CubeSat, American Family Insurance, Realtor.com, Zoku, Buffer, Thrive Smart\n  Systems, Topology Eyewear).\n- **The full method**, including the closing Mindset chapter (the eight experiment\n  pitfalls, leading through experimentation, and organizing/funding testing).\n- Prose descriptions are paraphrased summaries, not book text. Spot-checked numeric\n  thresholds (interview ranking accuracy, feature-stub funnel targets, brochure and\n  referral conversion targets) were verified against the book and confirmed.\n\nPage numbers point to the book for exact procedures and figures.\n\n## Contributing\n\nIssues and pull requests welcome, especially: tightening the coaching prompts in\n`SKILL.md` and fixing any place the summaries drift from the book. Keep the tone\ncoach-not-cheerleader and keep `SKILL.md` short.\n\n## License\n\nMIT, see [LICENSE](LICENSE). The skill implementation is MIT-licensed. The underlying\nmethodology, frameworks, and experiment names belong to David Bland, Alex\nOsterwalder, and Strategyzer, and are referenced here for education under fair use.\n\n## Privacy\n\nThis skill is plain text: no code execution, no network calls, no data\ncollection of any kind. See [PRIVACY.md](PRIVACY.md).\n",
  "bytes": 10624,
  "sha": "7820ff78d19f2139e265259fdc7104e9a2ea4fa60e3ccc48425c59ee5736205c",
  "repo_slug": "nimitbhargava/testing-business-ideas-skill",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_nimitbhargava_testing_business_ideas_ski_04de5232/readme"
}