Communitygithub.com

Capx-AI/casa

The validation gate. Turns a business idea into a validation kit and a GO or KILL verdict by running the validation playbooks. Use when validating an idea before committing to build a company.

casa란 무엇인가요?

casa is a Claude Code agent skill that the validation gate. Turns a business idea into a validation kit and a GO or KILL verdict by running the validation playbooks. Use when validating an idea before committing to build a company.

지원 대상~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/Capx-AI/casa/tree/HEAD/skills/casa-validate

즐겨 사용하는 AI에게 물어보기

이 에이전트 스킬이 미리 로드된 새 채팅을 엽니다.

문서

casa-validate

The wedge. Prove there is real demand before anyone builds anything. This is the validation gate every new idea passes through, and its verdict personalizes everything after it.

If company-brain/profile.json does not exist, tell the founder to run /casa-start first and stop. (casa-start invokes this gate itself for an idea-stage company.)

What it runs (the validation playbooks)

In dependency order (see each playbook's frontmatter under playbooks/level-0/):

  1. opportunity-scan (001): mine communities, reviews, search, trends for evidence of demand. Parallel-safe with 003, 004.
  2. problem-validation-interviews (002): Mom-Test discovery; problem evidence memo.
  3. competitive-teardown (003): map competitors, pricing, gaps.
  4. market-sizing-tam-sam-som (004): top-down and bottom-up; kill if the reachable market is below threshold.
  5. jobs-to-be-done-extraction (005): functional, emotional, social jobs.
  6. red-team-thesis (006): steelman why this fails; set explicit kill criteria.
  7. why-now-memo (007): the timing thesis.
  8. mvp-scoping (018): the thin slice to build, with explicit non-goals.

On a GO verdict, beachhead-selection (008, the first play of the next level) picks the narrowest wedge with the most compounding.

Output

  • A validation kit written to company-brain/ (evidence, sizing, beachhead, MVP scope).
  • A GO or KILL verdict against the red-team kill criteria.
  • On GO, a confirmed business profile (type, traits, ICP, monetization) for the router.

Rules

  • Evidence required. An opportunity is not valid until at least three independent signals from at least two channels support it. State confidence on every claim.
  • No fabrication. If a required input (an API, a real interview) is missing, say so and ask the founder rather than inventing data.
  • The landing-page demand test and live waitlist are part of this wedge but involve a real deploy and possibly paid actions; route those through explicit founder approval. For now, produce the kit and the verdict.
  • No em-dashes, no emojis.

관련 스킬