vannt-dev/aiws-factory
AIWS – AI Software Factory: turns a requirement into tested code through gated phases (analysis → design + test spec → human approval → per-task implementation with unit tests → review → PR). Language-agnostic, Claude Code first.
field: AI / ML workflow: Testing & TDDPlanning & specsFrontend & designPR / code reviewGit & commits
stack: JavaScript
ARCHETYPE
Fort Knox
Deny lists, pre-tool hooks, sandboxing. Nothing touches prod without a signature.
GUARDRAILS
4/5
Blocks destructive commands · Protects secrets · Pre-tool screening hook · No YOLO mode · details
Quality 19/100 Guardrails v2
Copy this rig
# review before running: this installs third-party code
$ npx degit vannt-dev/aiws-factory/.claude ./rig-aiws-factory # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit vannt-dev/aiws-factory/.claude/skills/api-design .claude/skills/api-design $ npx degit vannt-dev/aiws-factory/.claude/skills/be-conventions .claude/skills/be-conventions $ npx degit vannt-dev/aiws-factory/.claude/skills/coding-standards .claude/skills/coding-standards $ npx degit vannt-dev/aiws-factory/.claude/skills/db-design .claude/skills/db-design $ npx degit vannt-dev/aiws-factory/.claude/skills/fe-conventions .claude/skills/fe-conventions $ npx degit vannt-dev/aiws-factory/.claude/skills/legacy-analysis .claude/skills/legacy-analysis $ npx degit vannt-dev/aiws-factory/.claude/skills/requirement-analysis .claude/skills/requirement-analysis $ npx degit vannt-dev/aiws-factory/.claude/skills/unit-testing .claude/skills/unit-testing
$ curl -fsSL --create-dirs -o .claude/agents/analyst.md https://raw.githubusercontent.com/vannt-dev/aiws-factory/main/.claude/agents/analyst.md $ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/vannt-dev/aiws-factory/main/.claude/agents/architect.md $ curl -fsSL --create-dirs -o .claude/agents/developer.md https://raw.githubusercontent.com/vannt-dev/aiws-factory/main/.claude/agents/developer.md $ curl -fsSL --create-dirs -o .claude/agents/discovery.md https://raw.githubusercontent.com/vannt-dev/aiws-factory/main/.claude/agents/discovery.md $ curl -fsSL --create-dirs -o .claude/agents/planner.md https://raw.githubusercontent.com/vannt-dev/aiws-factory/main/.claude/agents/planner.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/vannt-dev/aiws-factory/main/.claude/agents/reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/test-designer.md https://raw.githubusercontent.com/vannt-dev/aiws-factory/main/.claude/agents/test-designer.md $ curl -fsSL --create-dirs -o .claude/agents/tester.md https://raw.githubusercontent.com/vannt-dev/aiws-factory/main/.claude/agents/tester.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"permissions": {
"deny": [
"Read(./**/.env)",
"Read(./**/.env.*)",
"Read(./**/secrets/**)",
"Read(./**/*.pem)",
"Read(./**/*.key)",
"Edit(./source-legacy/**)",
"Edit(./requirements/**)",
"Edit(./aiws/work/*/state.yaml)",
"Edit(./aiws/work/*/approvals/**)",
"Bash(aiws approve)",
"Bash(aiws approve *)",
"Bash(aiws reject)",
"Bash(aiws reject *)",
"Bash(aiws answer)",
"Bash(aiws answer *)",
"Bash(aiws redesign)",
"Bash(aiws redesign *)",
"Bash(aiws resume)",
"Bash(aiws resume *)",
"Bash(aiws unlock)",
"Bash(aiws unlock *)"
]
},
"hooks": {
"PreToolUse": [
{
"matcher": "Edit|Write|MultiEdit|NotebookEdit|Read",
"hooks": [
{
"type": "command",
"command": "node \"${CLAUDE_PROJECT_DIR}/aiws/adapters/cli/bin/aiws.js\" guard"
}
]
},
{
"matcher": "Bash|PowerShell",
"hooks": [
{
"type": "command",
"command": "node \"${CLAUDE_PROJECT_DIR}/aiws/adapters/cli/bin/aiws.js\" guard --bash"
}
]
}
]
}
} Skills (8)
api-designbe-conventionscoding-standardsdb-designfe-conventionslegacy-analysisrequirement-analysisunit-testing
Subagents (8)
| analyst model: opus | Phân tích một requirement thành mục tiêu, acceptance criteria đánh số, impact, reuse và câu hỏi mở. Viết aiws/work/<REQ>/01-analysis.md. |
| architect model: opus | Thiết kế giải pháp cho một REQ đã phân tích - quyết định chính, API, DB, FE/BE change, migration, rủi ro - và viết API contract OpenAPI. |
| developer model: opus | Hiện thực đúng MỘT task trong plan đã được duyệt, kèm unit test theo test spec. |
| discovery model: opus | Khảo sát source-fe, source-be, source-legacy và viết/cập nhật bản đồ hệ thống trong aiws/knowledge/. Dùng khi chạy `aiws discover` hoặc sau khi một REQ được merge. |
| planner model: opus | Chia design đã duyệt thành task nhỏ có danh sách file được phép sửa, test liên quan và phụ thuộc. Viết 04-plan.yaml. |
| reviewer model: opus | Review toàn bộ diff của nhánh REQ đối chiếu design, test spec và api-contract; ghi findings theo mức độ vào 05-review.md. |
| test-designer model: opus | Sinh test spec từ acceptance criteria và design, trước khi có code. Mỗi AC có ít nhất một test case. |
| tester model: sonnet | Chạy test của task/REQ và ghi kết quả vào evidence; không sửa code. Dùng ở chế độ tương tác - trong `aiws run` orchestrator tự chạy test. |
Hooks (2)
| event | matcher | runs |
|---|---|---|
| PreToolUse | Edit|Write|MultiEdit|NotebookEdit|Read | node "${CLAUDE_PROJECT_DIR}/aiws/adapters/cli/bin/aiws.js" guard |
| PreToolUse | Bash|PowerShell | node "${CLAUDE_PROJECT_DIR}/aiws/adapters/cli/bin/aiws.js" guard --bash |
Permissions
deny (21)
Read(./**/.env)
Read(./**/.env.*)
Read(./**/secrets/**)
Read(./**/*.pem)
Read(./**/*.key)
Edit(./source-legacy/**)
Edit(./requirements/**)
Edit(./aiws/work/*/state.yaml)
Edit(./aiws/work/*/approvals/**)
Bash(aiws approve)
Bash(aiws approve *)
Bash(aiws reject)
Bash(aiws reject *)
Bash(aiws answer)
Bash(aiws answer *)
Bash(aiws redesign)
Bash(aiws redesign *)
Bash(aiws resume)
Bash(aiws resume *)
Bash(aiws unlock)
Bash(aiws unlock *)
ask (0)
—
allow (0)
—
Similar rigs
FCabiddu/pocket-it
Agent system for Claude Code: analyst, architect, task-writer, developer, reviewer, security and tester, with hooks and scripts that enforce the gates
Orchestrator 3.2k tok · · Q5
vannt-dev/governed-agent-sdlc
Ship with agents. Keep humans in control. Cross-platform governance toolkit for AI coding agents.
Fort Knox 632 tok · · Q24
Zejnilovic/agd
Install, update and combine agentic setups from git repositories
Orchestrator 3.8k tok · · Q10
bpstack/harness
A reusable foundation for AI projects: rules, agents, templates, and commands.
Pragmatist 3.5k tok · · Q5