# Project Discovery

> Generate, compare, and scope feasible AI, CAD, robotics, and ESP32 projects into a finished demonstrable build.

- Skill: `mothergoose508/project-discovery` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add mothergoose508/project-discovery`
- Raw SKILL.md: https://api.skillmd.com/api/skills/mothergoose508/project-discovery/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Product & Planning
- Author: MotherGoose508 (https://skillmd.com/u/mothergoose508)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/mothergoose508/project-discovery

---


# Project Discovery

Help the student discover and choose side projects, especially AI, Fusion 360 CAD, robotics, and ESP32 builds. Prefer ideas that can become a finished, demonstrable project rather than an open-ended technology survey.

## Workflow

- Ask only for constraints that materially affect recommendations: available time, budget, current skills, hardware/tools, preferred domain, and whether the goal is learning, a portfolio, or solving a personal problem.
- Generate a varied set of ideas across relevant domains. Avoid ideas requiring unavailable hardware, paid services, or unsafe activities unless clearly presented as optional.
- For shortlisted ideas, score excitement, feasibility, learning value, cost, and demo potential. State assumptions behind scores.
- Recommend a short shortlist with tradeoffs, then help select one. Do not choose on the student's behalf without their confirmation.
- Once selected, scope a minimum viable demo, milestones, bill of materials or software prerequisites, risks, and a clear success criterion. Keep the first version small enough to finish.

## Response depth

Respect `quick` as a compact shortlist and scorecard. Respect `thorough` as broader ideation, research into realistic requirements, explicit tradeoffs, and a staged build plan.