Research Idea

**Judgment:** A strong, focused skill for Clawdbot users who ideate frequently. It’s well-suited for founders, indie hackers, and product people who want structured, repeatable analysis of business ideas directly from a Telegram chat. **Strengths:** - Very easy to trigger via `Idea: [description]` with no extra commands or tools. - Runs fully in the background using `sessions_spawn`, so it doesn’t block the main chat. - Produces a durable `research.md` file plus an in-chat summary and clear verdict (🟢/🟡/🟠/🔴). - Opinionated framework ensures coverage of concept, market, tech, business model, GTM, and risks. - No external CLIs or services, which improves reliability and simplifies installation. **Key Risks / Limitations:** - All “research” is model-generated; market sizes, competitor lists, and feasibility assessments may be incomplete, outdated, or incorrect if taken as factual. - Overly confident verdict emojis may encourage users to overweight the AI’s judgment vs. real-world validation. - Assumes a working Clawdbot environment with `sessions_spawn` and the specified directory structure; misconfiguration could break saving or returning results. - Single-template analysis: may be too rigid for highly specialized domains (e.g., deep biotech, complex regulation-heavy markets) without manual adjustment. **Recommended Scenarios:** - Early-stage ideation and triage: generating first-pass assessments before deeper manual research. - Comparing multiple candidate ideas to decide which deserve further validation or prototyping. - Creating initial research docs that can be refined by humans for pitch decks, product specs, or internal memos. - Use as a *starting point* for validation—paired with real customer interviews, market data, and expert review, not as a standalone decision-maker.