# Idea Evaluator

> Evaluate a raw app, startup, SaaS, internal tool, marketplace, or AI product idea before planning features. Use when the user shares a vague concept and needs pain analysis, market realism, feasibility, differentiation, risks, or a build/no-build recommendation.

- Skill: `saranskumar/idea-evaluator` (Agent Skill)
- Install (CLI): `npx skillmds@latest add saranskumar/idea-evaluator`
- Raw SKILL.md: https://api.skillmd.com/api/skills/saranskumar/idea-evaluator/raw
- Safety review: PASS (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: saranskumar (https://skillmd.com/u/saranskumar)
- Updated: 2026-08-19
- Page: https://skillmd.com/skills/saranskumar/idea-evaluator

---


# Goal
Turn a raw idea into a grounded evaluation that answers whether the idea is worth pursuing, reshaping, or rejecting.

# When to use
- The user presents an idea without a structured product thesis.
- A team wants to compare multiple ideas or wedges.
- A concept sounds exciting but may be weak on pain, distribution, or monetization.

# Instructions
1. Restate the idea in one sentence without hype.
2. Identify the primary user, painful job, current workaround, and trigger moment.
3. Score the idea from 1-5 on problem intensity, user clarity, urgency, distribution realism, monetization realism, feasibility, and defensibility seed.
4. Explain the strongest reason to build it and the strongest reason not to.
5. Recommend one of: proceed, reshape, validate first, or reject.
6. If the idea is vague, infer the most practical version and mark assumptions clearly.

# Constraints
- Do not confuse a broad market with a sharp initial wedge.
- Do not recommend building because the idea is technically interesting.
- Do not invent TAM numbers, user demand, or competitor facts without evidence.

# Output format
- One-line concept summary
- Evaluation table with scores and rationale
- Key risks
- Recommended wedge
- Decision and next action

# Examples
- "Evaluate an AI copilot for restaurant inventory ordering."
- "Is this B2B SaaS idea real or just founder fantasy?"
- "Compare these three internal tool ideas for an MVP."

# Output expectations
The answer should be decisive, specific, and honest enough to prevent wasted build effort.