Project Kickstart — Testing skill for Claude Code
Start any software project the right way with an AI agent: an adaptive interview, right-sized foundation docs (AGENTS.md, PRD, roadmap with test gates), and guided GitHub setup.
How to install Project Kickstart
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open motalim-ai/project-kickstart and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Project Kickstart does
Start any software project the right way with an AI agent: an adaptive interview, right-sized foundation docs (AGENTS.md, PRD, roadmap with test gates), and guided GitHub setup. Agent Skill + standalone prompt.
Alternatives in Testing
- Factory — Dark Factory Mode - Spec-in, software-out autonomous pipeline 2.8k ★
- Spec Kitty — Spec-Driven Development for serious software developers 1.6k ★
- Spec Driven Develop — Spec-driven development workflow for AI coding agents: architecture-first planning, task decomposition, GitHub 975 ★
README
Project Kickstart
**Start any software project the right way with an AI agent.**
An [Agent Skill](https://platform.claude.com/docs/en/agents-and-tools/agent-skills/overview) (and a standalone prompt for any AI agent) that interviews you about your project in your own language, writes the foundation docs your AI agents will follow, puts everything on GitHub, and explains what happens next.
[العربية](README.ar.md) · Made by **[Motalim](https://github.com/motalim-ai)**
Why
AI coding agents don't remember past sessions. Anything that isn't written down gets guessed, and every guess produces slightly different code. After a few weeks, that's spaghetti.
Project Kickstart fixes this before the first line of code. It writes down **what** you're building, **in what order**, **with which tools**, and **which rules** every agent must follow, including tests that must pass before moving to the next phase.
What it does
**Interviews you**, 3–5 simple questions at a time, in your language and dialect. It skips what doesn't apply (no SEO questions for an internal tool).
**Helps when you don't know.** Say "I don't know" and it explains the concept with an example from *your* project, recommends an option, and records the assumption so you can change it later.
**Looks before it asks:** your existing website, logo, designs or server (read-only, with permission).
**Writes a project brief** and waits for your approval.
**Picks a stack** and verifies the *current* versions at that moment, not from memory.
**Writes right-sized docs:**
Size Example Docs S script, prototype, simple landing page AGENTS.md,CLAUDE.md,README.md, short PRD, ROADMAPM app or site with users, database or admin panel + STACK, ARCHITECTURE, CODING-STANDARDS, TESTING, SECURITY, OPEN-QUESTIONS, ADRs L customers, payments, personal data, roles, public marketing + DATA-MODEL, DESIGN-SYSTEM, DEPLOYMENT,
Related Skills
Coding Interview Prep With Agents
Agent-guided coding interview prep for any role: training workflows, original exercises, Python test helpers,
Outlook LLM Tests
Run make crud-test from the repository root via the Bash tool. This is the primary, canonical way to execute t
Codetester
Automated test execution and guided manual testing. Reviews project docs, runs all automatable tests, then wal
Spec Driven Dev
A Claude Code skill that blocks code generation until a PRD and Technical Architecture Spec exist and are appr
Mobile UX
Uses a mobile-sized viewport to test the app for mobile UX
Raise PR
Raise a PR for the current feature branch — cleans up the spec folder, composes a PR body sized to the change,
Related Agents
Forge Plan
Forge's Planning agent — breaks a broad PRD or product brief into a phased roadmap, each phase sized to run th
Docs Reconciler
Drift detection agent for any software project. Use every 3–5 merged PRs, before a release, or when picking up
Preferences Setup
Sets up and updates job search preferences in config/preferences.yaml. Pass free-form text to extract preferen