project-architect
Documentation-first project planning agent skill. Generates specs, implementation plans, tasks, and single-shot prompts for coding agents. Compatible with Claude Code, Cursor, Codex, and 40+ agents via agentskills.io.
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Badge for your README
Markdown
[](/listings/ersinkoc--project-architect)HTML
<a href="/listings/ersinkoc--project-architect"><img src="/api/listings/ersinkoc--project-architect/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Universal (any agent) via skills CLI
The easiest way to install, works with all [agentskills.io-compatible agents](https://agentskills.io):
npx skills add ersinkoc/project-architect
This auto-detects your coding agent and installs to the correct location.
**Install for a specific agent:**
npx skills add ersinkoc/project-architect --agent wrongstack npx skills add ersinkoc/project-architect --agent claude-code npx skills add ersinkoc/project-architect --agent cursor npx skills add ersinkoc/project-architect --agent codex
**Install globally (all projects):**
npx skills add ersinkoc/project-architect --global
> Repository: [github.com/ersinkoc/project-architect](https://github.com/ersinkoc/project-architect)
Claude Code (plugin mode)
Add to your project's `.claude/settings.json`:
{
"plugins": [
"/absolute/path/to/project-architect"
]
}Or add to `~/.claude/settings.json` for global availability.
Manual installation (any agent)
Clone the repo and copy/symlink the skill directory into your agent's skills folder:
| Agent | Skills Directory | |-------|-----------------| | WrongStack | `.wrongstack/skills/` or `~/.wrongstack/skills/` | | Claude Code | `.claude/skills/` or `~/.claude/skills/` | | Cursor | `.cursor/skills/` or `~/.cursor/skills/` | | GitHub Copilot | `.github/skills/` | | Codex | `.codex/skills/` | | Gemini CLI | `.gemini/skills/` | | OpenCode | `.opencode/skills/` | | Generic | `.agents/skills/` or `~/.agents/skills/` |
# Example: install for WrongStack globally git clone https://github.com/ersinkoc/project-architect.git ln -s $(pwd)/project-architect ~/.wrongstack/skills/project-architect # Example: install for Claude Code globally ln -s $(pwd)/project-architect ~/.claude/skills/project-architect # Example: install for Cursor in a project ln -s $(pwd)/project-architect .cursor/skills/project-architect
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
plugin.json · Manifest · 8 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- {
- "name": "project-architect",
- "version": "1.1.0",
- "description": "Documentation-first project planning that produces agent-ready prompts. Generates interconnected docs (SPECIFICATION, IMPLEMENTATION, TASKS, BRANDING) plus a self-contained, single-shot PROMPT.md for execution by any coding agent (WrongStack, Claude Code, Cursor, and more).",
- "skills": [
- "SKILL.md"
- ]
- }
README.md · README · 246 lines · 2 findings marked
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- # Project Architect
- An [Agent Skill](https://agentskills.io) for **documentation-first project planning**. Transforms a project idea into implementation-ready blueprints and a single-shot coding agent prompt.
- Compatible with **40+ coding agents** including WrongStack, Claude Code, Cursor, Gemini CLI, GitHub Copilot, Codex, and more.
- ---
- ## ⚡ Sponsored by WrongStack
- <div align="center">
- ### _Built on the wrong stack. Shipped anyway._
- **[WrongStack](https://wrongstack.com)** is a free, [open-source](https://github.com/WrongStack/WrongStack) AI coding agent with a Brain, a Memory, and a full toolbox. It reads your code, edits files, runs commands, and coordinates specialist agents — across six surfaces, from a plain terminal REPL to a cross-machine HQ dashboard. No subscription required, and you keep your hand on every permission.
- [](https://wrongstack.com)
-
- [](https://github.com/WrongStack/WrongStack)
-
- [](https://github.com/WrongStack/WrongStack/stargazers)
- ```bash
- curl -fsSL https://wrongstack.com/install.sh | sh # macOS / Linux — self-contained binary
- Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
- Line 25 on GitHub
- ```
- ```powershell
- irm https://wrongstack.com/install.ps1 | iex # Windows (PowerShell) — no Node.js needed
- Note: Runs a downloaded PowerShell script (iex / Invoke-Expression)
- Line 29 on GitHub
- ```
- </div>
- | | What you get |
- |---|---|
- | 🌐 **[200+ LLM providers](https://wrongstack.com)** | Catalog pulled live from models.dev — Anthropic, OpenAI, Google, plus OAuth sign-in for Claude Pro/Max, ChatGPT and Copilot, and any OpenAI-compatible endpoint (Ollama, vLLM, LM Studio) |
- | 🛠️ **[67 built-in tools](https://github.com/WrongStack/WrongStack)** | Edits, lint/typecheck/test, execution, git, web, browser/E2E and a SQLite codebase index — every call gated by per-tool permissions |
- | 🧠 **[SAGE memory](https://github.com/WrongStack/WrongStack/blob/main/docs/sage/ARCHITECTURE.md)** | Project-wide long-term memory in SQLite/FTS5, anchored to files, symbols and commits — it gets better at *your* codebase over time |
- | 🖥️ **[Six surfaces](https://wrongstack.com)** | Readline REPL · Ink/React TUI (`--tui`) · WebUI · SimpleUI · Desktop · cross-machine HQ (`--hq`) |
- | 🤖 **[Fleet orchestration](https://github.com/WrongStack/WrongStack/blob/main/docs/director-architecture.md)** | A Director fans out specialist subagents over a project mailbox; `eternal` & `parallel` goal loops run until the contract verifies |
- | 🔍 **[Chimera & Kanban](https://wrongstack.com)** | Auto-review agents that critique your diffs with severity-ranked findings, plus durable Kanban boards with atomic verification |
- | 🔐 **[Secure by default](https://github.com/WrongStack/WrongStack/blob/main/SECURITY.md)** | Encrypted secrets at rest, project-root containment, opt-in YOLO mode — MIT licensed, TypeScript-strict |
- > **📊 The perfect pairing:** This band tells you exactly where your Claude limits stand — and **WrongStack** keeps you moving when they close in. It reads plan windows for Claude, ChatGPT, Copilot, Z.AI and more right in its own statusline and quota page, and when one provider runs dry, **fallback chains** rotate you onto the next model automatically. Watch the band, dodge the wall.
- <div align="center">
- 🔗 **[wrongstack.com](https://wrongstack.com)** · **[github.com/WrongStack/WrongStack](https://github.com/WrongStack/WrongStack)** · ⭐ **[Star it on GitHub](https://github.com/WrongStack/WrongStack/stargazers)**
- </div>
- ---
- ## What It Does
- Given a project idea, Project Architect walks you through an interactive discovery process and generates 5 interconnected documents:
- ```
- [Discovery] -> SPECIFICATION.md -> IMPLEMENTATION.md -> TASKS.md -> BRANDING.md
- (The What) (The How) (The Work) (Identity)
- | | |
- +-------------------+------------------+
- |
- PROMPT.md
- (Single-Shot Agent Prompt)
- ```
- | Document | Purpose |
- |----------|---------|
- | **SPECIFICATION.md** | What the project is, features, data model, API surface |
- | **IMPLEMENTATION.md** | Tech stack, design patterns, directory structure, schemas |
- | **TASKS.md** | Ordered work items, each completable in a single agent session |
- | **BRANDING.md** | Name, colors, typography, voice (optional, for user-facing projects) |
- | **PROMPT.md** | Self-contained prompt to build the entire project from scratch |
- ## Features
- - **Interactive tech stack advisor** -- presents options with trade-offs, lets you choose
- - **Design pattern recommendations** -- matched to your project's specific needs with code sketches
- - **Scales to project size** -- weekend hack gets 15 tasks, enterprise system gets 100+
- - **Agent-optimized output** -- every task lists exact files to create/modify with acceptance criteria
- - **Pause-and-review flow** -- generates each document, waits for approval before continuing
- ## Installation
- ### Universal (any agent) via skills CLI
- The easiest way to install, works with all [agentskills.io-compatible agents](https://agentskills.io):
- ```bash
- npx skills add ersinkoc/project-architect
- ```
- This auto-detects your coding agent and installs to the correct location.
- **Install for a specific agent:**
- ```bash
- npx skills add ersinkoc/project-architect --agent wrongstack
- npx skills add ersinkoc/project-architect --agent claude-code
- npx skills add ersinkoc/project-architect --agent cursor
- npx skills add ersinkoc/project-architect --agent codex
- ```
- **Install globally (all projects):**
- ```bash
- npx skills add ersinkoc/project-architect --global
- ```
- > Repository: [github.com/ersinkoc/project-architect](https://github.com/ersinkoc/project-architect)
- ### Claude Code (plugin mode)
- Add to your project's `.claude/settings.json`:
- ```json
- {
- "plugins": [
- "/absolute/path/to/project-architect"
- ]
- }
- ```
- Or add to `~/.claude/settings.json` for global availability.
- ### Manual installation (any agent)
- Clone the repo and copy/symlink the skill directory into your agent's skills folder:
- | Agent | Skills Directory |
- |-------|-----------------|
- | WrongStack | `.wrongstack/skills/` or `~/.wrongstack/skills/` |
- | Claude Code | `.claude/skills/` or `~/.claude/skills/` |
- | Cursor | `.cursor/skills/` or `~/.cursor/skills/` |
- | GitHub Copilot | `.github/skills/` |
- | Codex | `.codex/skills/` |
- | Gemini CLI | `.gemini/skills/` |
- | OpenCode | `.opencode/skills/` |
- | Generic | `.agents/skills/` or `~/.agents/skills/` |
- ```bash
- # Example: install for WrongStack globally
- git clone https://github.com/ersinkoc/project-architect.git
- ln -s $(pwd)/project-architect ~/.wrongstack/skills/project-architect
- # Example: install for Claude Code globally
- ln -s $(pwd)/project-architect ~/.claude/skills/project-architect
- # Example: install for Cursor in a project
- ln -s $(pwd)/project-architect .cursor/skills/project-architect
- ```
- ## Usage
- Start a conversation with your coding agent and describe what you want to build:
- ```
- > plan my project: a CLI tool for managing dotfiles across machines
- ```
- Or use any of these trigger phrases:
- - "plan my project"
- - "spec this out"
- - "architect a system for..."
- - "help me plan"
- - "what stack should I use"
- - "generate a prompt"
- - "break this into tasks"
- - "I want to build X"
- ## Workflow
- 1. **Discovery** -- Agent asks structured questions about your project (type, scope, stack, features)
- 2. **SPECIFICATION.md** -- Generated and presented for review
- 3. **IMPLEMENTATION.md** -- Tech decisions, patterns, directory structure
- 4. **TASKS.md** -- Ordered work breakdown with file lists and acceptance criteria
- 5. **BRANDING.md** -- Optional identity guide (colors, typography, voice)
- 6. **PROMPT.md** -- Everything synthesized into a single executable prompt
- You review and approve each document before the next is generated.
- ## Partial Workflows
- You don't have to run the full pipeline:
- | What You Say | What Happens |
- |-------------|-------------|
- | "Just the spec" | Generates SPECIFICATION.md only |
- | "Skip to tasks" | Lightweight spec + impl, then detailed tasks |
- | "Just give me a prompt" | Condensed discovery, straight to PROMPT.md |
- | "Help me choose a stack" | Interactive tech stack selection only |
- | "What patterns should I use?" | Design pattern consultation |
- ## Reference Files
- The skill includes detailed guides that inform each generation phase:
- | File | Purpose |
- |------|---------|
- | `references/elicitation-guide.md` | Question framework for project discovery |
- | `references/tech-stacks.md` | Interactive tech stack selection with trade-offs |
- | `references/design-patterns.md` | Pattern catalog with selection guide |
- | `references/specification-guide.md` | Template and rules for SPECIFICATION.md |
- | `references/implementation-guide.md` | Template and rules for IMPLEMENTATION.md |
- | `references/tasks-guide.md` | Template and rules for TASKS.md |
- | `references/branding-guide.md` | Template and rules for BRANDING.md |
- | `references/agent-prompt.md` | Template and rules for PROMPT.md |
- ## Project Structure
- ```
- project-architect/
- ├── SKILL.md # Skill definition (agentskills.io format)
- ├── plugin.json # Claude Code plugin manifest
- ├── LICENSE # MIT License
- ├── README.md # This file
- └── references/
- ├── elicitation-guide.md # Discovery question framework
- ├── tech-stacks.md # Tech stack advisor
- ├── design-patterns.md # Pattern catalog
- ├── specification-guide.md # Spec template
- ├── implementation-guide.md # Implementation template
- ├── tasks-guide.md # Tasks template
- ├── branding-guide.md # Branding template
- └── agent-prompt.md # Prompt template
- ```
- ## Compatibility
- This skill follows the [Agent Skills specification](https://agentskills.io/specification) and is compatible with:
- - **WrongStack** (terminal-native AI coding agent — [wrongstack.com](https://wrongstack.com))
- - **Claude Code** (plugin + skill)
- - **Cursor**
- - **GitHub Copilot**
- - **OpenAI Codex**
- - **Gemini CLI**
- - **OpenCode**
- - **Kiro**
- - **And 30+ more agents** via [agentskills.io](https://agentskills.io)
- ## License
- MIT
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.