create-harness-vibe-coding 0.7.2 → 0.8.0

package/README.md

@@ -1,210 +1,233 @@
-<p align="center">
-  <img src="https://img.shields.io/npm/v/create-harness-vibe-coding?color=blue" alt="npm">
-  <img src="https://img.shields.io/badge/node-%3E%3D18-brightgreen" alt="node">
-  <img src="https://img.shields.io/npm/l/create-harness-vibe-coding" alt="license">
-  <img src="https://img.shields.io/github/stars/zingspark/create-harness-vibe-coding?style=social" alt="stars">
-</p>
-
-<h1 align="center">create-harness-vibe-coding</h1>
-<p align="center">
-  <b>A harness for your AI agent. One scaffold. Zero drift.</b><br>
-</p>
-
-## One Command. Done.
-
-```bash
-npx create-harness-vibe-coding@latest my-project
-```
-
-## Your Agent Knows What to Do
-
-Already have a project? **Don't read the docs**. Paste this sentence. Your agent handles the rest.
-
-```text
-Follow the README at https://github.com/zingspark/create-harness-vibe-coding to configure this project with create-harness-vibe-coding; before editing, ask the Agent-link install intake questions; for a new project run the 0-1 bootstrap, and for an existing project or legacy architecture run a dry-run first, preserve existing files, merge only missing Harness guidance, then follow Harness/SETUP.md.
-```
-
-That's it. Two paths into the harness — you type `npx`, or your agent reads the sentence.
-
-Chinese README: [README-CN.md](README-CN.md)
-
----
-
-## What You Get
-
-| You get | So your agent |
-|---------|---------------|
-| `CLAUDE.md` + `Harness/README.md` | Starts with a router, not a novel |
-| `Harness/tasks/` + `Harness/PROGRESS.md` | Tracks work across sessions |
-| `/wf` workflow + heartbeat | Finishes long tasks without getting lost |
-| `/wf update` | Pulls scaffold fixes from GitHub |
-| `subagent-orchestrator` | Runs parallel agents without collision |
-| `memory-master` + `context-master` | Learns from failures, compresses when full |
-| PRD + Research templates | Asks "what" and "why" before coding |
-| 11 built-in agents | Research, plan, architect, test, build, review, debug, verify |
-| Architecture docs | Knows where boundaries live |
-| Context-loading protocol | Loads only the docs each agent needs |
-| `.claude/` skeleton | Agents, skills, commands, hooks — ready to go |
-
----
-
-## Why This Exists
-
-Most AI coding projects fail before anyone writes a line of bad code. The agent jumps straight to implementation, drifts from intent, forgets yesterday's decisions, and bloats its context with the whole repo.
-
-| Without harness | With harness |
-|-----------------|--------------|
-| Idea → code. Hope. | Idea → Research → PRD → Architecture → Build → Verify |
-| Agent reads everything | Router loads the one doc it needs |
-| Subagent gets a vague "fix it" | Context pack: role, boundary, return format |
-| Drift invisible until demo | Validator flags missing pieces |
-| Long task stalls, context explodes | `/wf` heartbeat + recovery loop |
-| Scaffold rots | `/wf update` pulls latest from GitHub |
-
----
-
-## How It Works
-
+<p align="center">
+  <img src="https://img.shields.io/npm/v/create-harness-vibe-coding?color=blue" alt="npm">
+  <img src="https://img.shields.io/badge/node-%3E%3D18-brightgreen" alt="node">
+  <img src="https://img.shields.io/npm/l/create-harness-vibe-coding" alt="license">
+  <img src="https://img.shields.io/github/stars/zingspark/create-harness-vibe-coding?style=social" alt="stars">
+</p>
+
+<h1 align="center">create-harness-vibe-coding</h1>
+<p align="center">
+  <b>A harness for your AI agent. One scaffold. Zero drift.</b><br>
+</p>
+
+## One Command. Done.
+
+```bash
+npx create-harness-vibe-coding@latest my-project
+```
+
+## Your Agent Knows What to Do
+
+Already have a project? **Don't read the docs**. Paste this sentence. Your agent handles the rest.
+
 ```text
-npx create-harness-vibe-coding@latest my-project
-    ↓
-Agent reads Harness/SETUP.md
-    ↓
-Router loads only what the task needs
-    ↓
-PRD → Research → Architecture → first task capsule
-    ↓
-Build → Test → Review → Verify → Feedback
-    ↓
-/wf update keeps the harness current
-```
-
-```mermaid
-graph TD
-    A[ /wf Enter ] --> B[ Explore: 3+ parallel agents ]
-    B --> C[ Second Plan ]
-    C --> D[ Build: test → implement ]
-    D --> E[ Dual-Gate Review ]
-    E --> F{ Pass? }
-    F -->| No | G[ Debug → Fix → Loop ]
-    G --> E
-    F -->| Yes | H[ Closeout: memory + context ]
-    H --> I[ /wf update ]
-```
-
----
-
-## Usage
-
-### New project
-
-```bash
-npx create-harness-vibe-coding@latest my-project
-cd my-project
-# Your agent reads Harness/SETUP.md. Done.
+Follow the README at https://github.com/zingspark/create-harness-vibe-coding to configure this project with create-harness-vibe-coding; before editing, scan the project root and ask the Agent-link install intake questions; if Harness/ already exists, ask how to handle it before writing; for a new project run the 0-1 bootstrap, and for an existing project or legacy architecture run a dry-run first, preserve existing files, merge only missing Harness guidance, then follow Harness/SETUP.md.
 ```
-
+
+That's it. Two paths into the harness — you type `npx`, or your agent reads the sentence.
+
+Chinese README: [README-CN.md](README-CN.md)
+
+---
+
+## What You Get
+
+| You get | So your agent |
+|---------|---------------|
+| `CLAUDE.md` + `Harness/README.md` | Starts with a router, not a novel |
+| `Harness/tasks/` + `Harness/PROGRESS.md` | Tracks work across sessions |
+| `/wf` / `$wf` workflow + heartbeat | Finishes long tasks without getting lost |
+| `/wf-update` / `$wf-update` | Pulls scaffold fixes from GitHub |
+| `subagent-orchestrator` | Runs parallel agents without collision |
+| `memory-master` + `context-master` | Learns from failures, compresses when full |
+| PRD + Research templates | Asks "what" and "why" before coding |
+| 11 built-in agents | Research, plan, architect, test, build, review, debug, verify |
+| Architecture docs | Knows where boundaries live |
+| Context-loading protocol | Loads only the docs each agent needs |
+| `.claude/` + `.agents/skills/` | Claude Code and Codex skill adapters over the same Harness docs |
+
+---
+
+## Why This Exists
+
+Most AI coding projects fail before anyone writes a line of bad code. The agent jumps straight to implementation, drifts from intent, forgets yesterday's decisions, and bloats its context with the whole repo.
+
+| Without harness | With harness |
+|-----------------|--------------|
+| Idea → code. Hope. | Idea → Research → PRD → Architecture → Build → Verify |
+| Agent reads everything | Router loads the one doc it needs |
+| Subagent gets a vague "fix it" | Context pack: role, boundary, return format |
+| Drift invisible until demo | Validator flags missing pieces |
+| Long task stalls, context explodes | `/wf` heartbeat + recovery loop |
+| Scaffold rots | `/wf-update` or `$wf-update` pulls latest from GitHub |
+
+---
+
+## How It Works
+
+```text
+npx create-harness-vibe-coding@latest my-project
+    ↓
+Agent reads Harness/SETUP.md
+    ↓
+Router loads only what the task needs
+    ↓
+PRD → Research → Architecture → first task capsule
+    ↓
+Build → Test → Review → Verify → Feedback
+    ↓
+/wf-update or $wf-update keeps the harness current
+```
+
+```mermaid
+graph TD
+    A[ /wf Enter ] --> B[ Explore: 3+ parallel agents ]
+    B --> C[ Second Plan ]
+    C --> D[ Build: test → implement ]
+    D --> E[ Dual-Gate Review ]
+    E --> F{ Pass? }
+    F -->| No | G[ Debug → Fix → Loop ]
+    G --> E
+    F -->| Yes | H[ Closeout: memory + context ]
+    H --> I[ /wf-update or $wf-update ]
+```
+
+---
+
+## Usage
+
+### New project
+
+```bash
+npx create-harness-vibe-coding@latest my-project
+cd my-project
+# Your agent reads Harness/SETUP.md. Done.
+```
+
 ### Existing project — safe merge
 
 ```bash
 # Preview first. Always.
 npx create-harness-vibe-coding@latest my-app . -y --dry-run
-
+
 # Add only what's missing. Never overwrite.
 npx create-harness-vibe-coding@latest my-app . -y --on-conflict skip
 ```
 
-| Flag | Does |
-|------|------|
-| `-y` | Skip prompts |
-| `--dry-run` | Preview — no writes |
-| `--on-conflict skip` | Keep your files, add only new ones |
-| `--on-conflict backup` | Rename existing → write new |
-| `--on-conflict overwrite` | Replace (destructive) |
-| `--list-options` | Show optional workflows |
+`npx` is an install and safe-merge entry, not an update engine for an already installed Harness. Once `Harness/` exists, use `/wf-update` in Claude Code, `$wf-update` in Codex, or `node Harness/scripts/wf-update-check.mjs`; root entry conflicts such as `CLAUDE.md`, `AGENTS.md`, `.claude/`, `.agents/`, `.codex/`, and local Harness docs need agent-mediated merge decisions.
+
+| Flag | Does |
+|------|------|
+| `-y` | Skip prompts |
+| `--dry-run` | Preview — no writes |
+| `--on-conflict skip` | Keep your files, add only new ones |
+| `--on-conflict backup` | Rename existing → write new |
+| `--on-conflict overwrite` | Replace (destructive) |
+| `--list-options` | Show optional workflows |
 | `--with <ids>` | Add workflow by id |
+| `--recommend <ids>` | Record recommendation-only external capabilities |
 | `--preset <name>` | Add `web-app` or `fullstack` preset |
 
-### Optional workflows
-
-```bash
+Without `-y`, interactive mode offers checkbox selection for optional workflows and external recommendations.
+
+### Optional workflows
+
+```bash
 npx create-harness-vibe-coding@latest my-app -y --with browser-e2e
 npx create-harness-vibe-coding@latest my-app -y --preset web-app
+npx create-harness-vibe-coding@latest my-app -y --recommend superpowers,codegraph
 ```
-
-| Workflow | For |
-|----------|-----|
-| `browser-e2e` | Screenshots, traces, smoke tests |
-| `ui-ux-review` | Responsive, a11y, polish |
-| `ts-react-frontend` | TypeScript + React + Vite |
+
+| Workflow | For |
+|----------|-----|
+| `browser-e2e` | Screenshots, traces, smoke tests |
+| `ui-ux-review` | Responsive, a11y, polish |
+| `ts-react-frontend` | TypeScript + React + Vite |
 | `python-backend` | FastAPI, pytest |
 | `github-pr-review` | PR diff review + CI evidence |
 
-### Agent-link intake
-
-When your agent reads the one-sentence prompt above, it asks **at most 3 questions** before touching files:
-
-- Is `CLAUDE.md` or `AGENTS.md` already there? → merge, don't replace
-- Is `docs/` used for product docs? → puts harness in `Harness/`, not `docs/`
-- What stack? → installs matching optional workflows
-
-If a file already exists, the agent asks first. The default is always **preserve**.
+External recommendations are recorded in `Harness/SETUP.md` but not installed automatically:
 
-### After scaffolding
+| Recommendation | For |
+|----------------|-----|
+| `superpowers` | Community skill registry and agent workflows |
+| `caveman` | Strict low-abstraction workflow/style capability |
+| `agent-research` | Broader product, dependency, and ecosystem research agents |
+| `codegraph` | Code graph or repository-map tooling |
 
-```text
-"Read Harness/SETUP.md. Bootstrap this project."
-"Use /wf for this migration."
-"/wf update — pull latest harness improvements."
-```
+### Agent-link intake
 
-### Verify
+When your agent reads the one-sentence prompt above, it must first scan the project root, then ask **at most 3 blocking questions** before touching files.
 
-```bash
-npm test
-node Harness/scripts/validate-harness.mjs
-```
+Root scan checklist: list top-level files, detect `CLAUDE.md`, `AGENTS.md`, `.claude/`, `.agents/`, `.codex/`, `Harness/`, package files, CI files, docs folders, and obvious app entry points.
 
----
+Ask these only when they affect writes:
 
-## Inside
+- If `CLAUDE.md` or `AGENTS.md` already exists, should missing Harness guidance be merged, kept separate, or skipped?
+- If `Harness/` already exists, should the agent run update/dry-run, merge missing files, keep it untouched, or remove/reinstall after approval?
+- Which optional capabilities should be enabled: local workflows (`browser-e2e`, `ui-ux-review`, `ts-react-frontend`, `python-backend`, `github-pr-review`) and recommendation-only external options (`Superpowers`, `Caveman`, `agent research`, `code graph`)?
 
-```
-my-project/
-├── CLAUDE.md                  ← Agent entry
-├── AGENTS.md                  ← Agent registry
-├── .gitignore
-├── Harness/
-│   ├── README.md              ← Doc router
-│   ├── SETUP.md               ← Bootstrap guide (delete after init)
-│   ├── MEMORY.md              ← Resource index
-│   ├── PROGRESS.md            ← Task tracker
-│   ├── WF.md / WF-MAX.md      ← Workflow modes
-│   ├── tasks/                 ← Per-task capsules
-│   ├── research/              ← PRD + research templates
-│   ├── memory/                ← Durable self-learning
-│   └── scripts/               ← Validator
+Harness docs always live in root `Harness/`; do not route Harness files through `docs/`. If a file already exists, the agent asks first. The default is always **preserve**.
+
+### After scaffolding
+
+```text
+"Read Harness/SETUP.md. Bootstrap this project."
+"Use /wf for this migration."          # Claude Code
+"Use $wf for this migration."          # Codex
+"/wf-update" or "$wf-update"           # pull latest harness improvements
+```
+
+### Verify
+
+```bash
+npm test
+node Harness/scripts/validate-harness.mjs
+```
+
+---
+
+## Inside
+
+```
+my-project/
+├── CLAUDE.md                  ← Agent entry
+├── AGENTS.md                  ← Agent registry
+├── .gitignore
+├── Harness/
+│   ├── README.md              ← Doc router
+│   ├── SETUP.md               ← Bootstrap guide (delete after init)
+│   ├── MEMORY.md              ← Resource index
+│   ├── PROGRESS.md            ← Task tracker
+│   ├── WF.md / WF-MAX.md      ← Workflow modes
+│   ├── tasks/                 ← Per-task capsules
+│   ├── research/              ← PRD + research templates
+│   ├── memory/                ← Durable self-learning
+│   └── scripts/               ← Validator
 ├── .claude/
 │   ├── agents/               ← 11 common agents
-│   ├── skills/               ← Harness loaders
-│   ├── commands/             ← /wf, /wf update
+│   ├── skills/               ← Claude Code skill commands
 │   └── rules/                ← Universal coding rules
+├── .agents/
+│   └── skills/               ← Codex repo skills
+├── .codex/
+│   └── hooks.json             ← Codex hook bridge
 └── tests/
 ```
 
-`Harness/` holds all harness docs. `.claude/` stays at root — that's where Claude Code discovers agents, skills, and commands.
-
----
-
-## Footprint
-
-| | |
-|---|---|
-| Runtime | None |
-| Dependencies | 2 (`@clack/prompts`, `picocolors`) |
-| Node | ≥ 18 |
-| Generated code | None until you pick a stack |
-
----
-
-MIT © [zingspark](https://github.com/zingspark)
+`Harness/` holds all harness docs. Tool-discovery folders stay at root: `.claude/` for Claude Code, `.agents/skills/` for Codex repo skills, and `.codex/` for Codex config/hooks.
+
+---
+
+## Footprint
+
+| | |
+|---|---|
+| Runtime | None |
+| Dependencies | 2 (`@clack/prompts`, `picocolors`) |
+| Node | ≥ 18 |
+| Generated code | None until you pick a stack |
+
+---
+
+MIT © [zingspark](https://github.com/zingspark)

package/bin/create-harness-vibe-coding.js

@@ -1,2 +1,2 @@
-#!/usr/bin/env node
-import('../src/index.js');
+#!/usr/bin/env node
+import('../src/index.js');

package/package.json

@@ -1,43 +1,45 @@
-{
-  "name": "create-harness-vibe-coding",
-  "version": "0.7.2",
-  "description": "Scaffold a 0-1 product harness for AI-assisted research, PRD, planning, architecture, build, test, and feedback loops",
-  "type": "module",
-  "bin": {
-    "create-harness-vibe-coding": "./bin/create-harness-vibe-coding.js"
-  },
-  "files": [
-    "bin/",
-    "src/",
-    "templates/",
-    "README-CN.md"
-  ],
-  "scripts": {
-    "start": "node src/index.js",
-    "build:version": "node scripts/build-version.mjs",
-    "test": "node --test tests/*.test.js",
-    "test:smoke": "node --test tests/cli-smoke.test.js",
-    "pack:smoke": "node --test tests/pack-smoke.test.js"
-  },
-  "dependencies": {
-    "@clack/prompts": "^0.7.0",
-    "picocolors": "^1.1.0"
-  },
-  "engines": {
-    "node": ">=18"
-  },
-  "repository": {
-    "type": "git",
-    "url": "https://github.com/zingspark/create-harness-vibe-coding.git"
-  },
-  "keywords": [
-    "claude-code",
-    "harness",
-    "scaffold",
-    "vibe-coding",
-    "agentic",
-    "ecc",
-    "superpowers"
-  ],
-  "license": "MIT"
-}
+{
+  "name": "create-harness-vibe-coding",
+  "version": "0.8.0",
+  "description": "Scaffold a 0-1 product harness for AI-assisted research, PRD, planning, architecture, build, test, and feedback loops",
+  "type": "module",
+  "bin": {
+    "create-harness-vibe-coding": "bin/create-harness-vibe-coding.js"
+  },
+  "files": [
+    "bin/",
+    "src/",
+    "templates/",
+    "README-CN.md"
+  ],
+  "scripts": {
+    "start": "node src/index.js",
+    "build:version": "node scripts/build-version.mjs",
+    "prepush": "node scripts/pre-push-check.mjs",
+    "prepush:install": "node scripts/pre-push-check.mjs --install",
+    "test": "node --test tests/*.test.js",
+    "test:smoke": "node --test tests/cli-smoke.test.js",
+    "pack:smoke": "node --test tests/pack-smoke.test.js"
+  },
+  "dependencies": {
+    "@clack/prompts": "^0.7.0",
+    "picocolors": "^1.1.0"
+  },
+  "engines": {
+    "node": ">=18"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/zingspark/create-harness-vibe-coding.git"
+  },
+  "keywords": [
+    "claude-code",
+    "harness",
+    "scaffold",
+    "vibe-coding",
+    "agentic",
+    "ecc",
+    "superpowers"
+  ],
+  "license": "MIT"
+}