loop-engineering
Run the repository's end-to-end delivery loop from idea or legacy-project audit to plan, implementation, verification, documentation, and release preparation.
Skill by NoMoneyDaddy
Chalk Labs
Need help deploying loop-engineering?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/nomoneydaddy-loop-engineering-d5a7f3
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: security
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions: not checked yet.
What's inside
- Skill name
loop-engineering- Description
- Run the repository's end-to-end delivery loop from idea or legacy-project audit to plan, implementation, verification, documentation, and release preparation. Use when the user wants structured interactive guidance followed by persistent execution with minimal unnecessary stops.
- License
- MIT
- Instructions
- 154 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# loop-engineering
本 skill 是專案主工作流。
## 讀取順序
1. `AGENTS.md`
2. `docs/session_loop_contract.md`
3. `docs/interactive_project_flow.md`
4. `docs/project_architecture_best_practices.md`
5. `docs/project_lifecycle_automation.md`
6. `docs/subagent_dispatch.md`
## 兩種模式
### A. 新專案
1. 必要時先用 `interview-me`
2. 用 `idea-refine` 收斂想法
3. 用 `spec-driven-development` 建 spec
4. 技術棧定了就先用 `project-config-generation`
5. 用 `planning-and-task-breakdown` 切 task
6. 再進實作 loop
### B. 舊專案改善
1. 先看入口檔、設定、測試、部署
2. 先提改善與重構切片
3. 使用者確認後才做大改
## 預設流程
這是預設路徑,不是固定步數。
1. 先判斷現在是新專案還是舊專案
2. 只問 1-3 組決策題
3. 先跑 `node scripts/inspect_agent_capabilities.js`
4. 先讀:
- `docs/loop_maturity_model.md`
- `docs/engineering_phase_loop.md`
- `docs/loop_circuit_breaker.md`
5. 根據情境讀:
- `prompts/interactive_tech_stack_prompt.md`
- `prompts/interactive_api_key_prompt.md`
- `prompts/legacy_project_improvement_prompt.md`
6. 若缺 `.loop/*`,先用 `goal-loop` 或 `node scripts/init_session_loop.js . --goal "<objective>"`
7. 技術棧確認後,先用 `project-config-generation` 落設定檔
8. 先給短版方案
9. 視任務大小決定:
- 小任務:直接切 task 開工
- 中大型任務:再補完整計畫與 task 切片
10. 完成必要確認後,進入 build-test-fix-doc loop
11. 每輪記錄 `maturity_level`、`retry_count`、`blocked_reason`、`convergence_state`
12. 若是 JS / TS 專案且有 Biome,啟用 quality loop:
- `npx @biomejs/biome check --write <changed-paths>`
- 必要時再跑測試
13. 進 merge / ship 前補:
- `code-review-and-quality`
- `observability-and-instrumentation`
- `shipping-and-launch`
## 初始化
若是新專案或缺少核心檔案,先執行:
```bash
node scripts/init_project_workspace.js <targetDir> --name <projectName> --idea "<idea>"
```
再依情境推薦 skills / tools:
```bash
node scripts/auto_skill_setup.js --project-type <type> --ui-style <style> --deployment <target> --language <lang> --database <db>
```
## 實作 loop
每輪都做:
1. 選一個最小 task
2. 最小必要修改
3. 跑對應驗證
4. 失敗就修根因
5. 看 circuit breaker 是否該停
6. 更新:
- `.loop/PLAN.md`
- `.loop/STATE.json`
- `.loop/CHECKPOINTS.md`
- `.loop/EVIDENCE.md`
- 必要時記 `tier`、`orientation`、`time_budget`
7. 若流程穩定可重用,參考 `docs/skill_crystallization_loop.md`
8. 更新:
- `docs/TASKS.md`
- `docs/DEBUG_NOTES.md`
- `docs/STATE.md`
- `README.md`
## 技能路由
- 想法模糊:`idea-refine`
- 要 `/goal` 類跨平台 loop:`goal-loop`
- 缺 spec:`spec-driven-development`
- 工作太大:`planning-and-task-breakdown`
- 技術棧已定但設定檔未落地:`project-config-generation`
- 開始實作:`incremental-implementation`
- 行為變動:`test-driven-development`
- framework 問題:`source-driven-development`
- 前端:`frontend-ui-engineering`
- API 邊界:`api-and-interface-design`
- runtime 驗證:`browser-testing-with-devtools`
- 除錯:`debugging-and-error-recovery`
- JS / TS 品質 loop:`biome-quality-automation`
- 合併前:`code-review-and-quality`
- 上線前觀測:`observability-and-instrumentation`
- 收尾文件:`documentation-and-adrs`
- 發版:`shipping-and-launch`
## 工具路由From NoMoneyDaddy/demo-app, MIT.
More from NoMoneyDaddy/demo-app
- 0.474score
planning-and-task-breakdown
NoMoneyDaddy
Break a spec into small verifiable tasks with clear acceptance criteria.
SkillCFreshness: CClaude Code +9 - 0.469score
incremental-implementation
NoMoneyDaddy
Implement work in thin vertical slices that keep the system runnable and testable after each step.
SkillCFreshness: CClaude Code +9 - 0.469score
goal-loop
NoMoneyDaddy
Run a cross-platform session loop using repo files instead of platform-specific slash commands.
SkillCFreshness: CClaude Code +9 - 0.469score
interview-me
NoMoneyDaddy
Run a one-question-at-a-time interview to extract what the user actually needs before planning or implementation.
SkillCFreshness: CClaude Code +9 - 0.464score
api-and-interface-design
NoMoneyDaddy
Design stable, explicit, hard-to-misuse interfaces across modules, APIs, services, and data contracts.
SkillCFreshness: CClaude Code +9 - 0.464score
code-review-and-quality
NoMoneyDaddy
Review changes across correctness, readability, architecture, security, and performance before merge.
SkillCFreshness: CClaude Code +9 - 0.464score
observability-and-instrumentation
NoMoneyDaddy
Add logs, metrics, tracing, and runtime signals so shipped systems are diagnosable.
SkillCFreshness: CClaude Code +9 - 0.464score
documentation-and-adrs
NoMoneyDaddy
Capture important decisions, project state, and maintenance-facing documentation as part of delivery.
SkillCFreshness: CClaude Code +9 - 0.464score
performance-optimization
NoMoneyDaddy
Optimize performance only after measuring and locating the bottleneck.
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jun 26, 2026 (4 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/NoMoneyDaddy/demo-app
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Same skill elsewhere
Copies of this SKILL.md published by others; search shows only one of them.
More from NoMoneyDaddy
- 0.479score
loop-engineering
NoMoneyDaddy
Run the repository's end-to-end delivery loop from idea or legacy-project audit to plan, implementation, verification, documentation, and release preparation.
SkillCFreshness: CClaude Code +9 - 0.474score
cavekit-spec
NoMoneyDaddy
Maintain `docs/SPEC.md` as the structured spec source of truth.
SkillCFreshness: CClaude Code +9 - 0.474score
planning-and-task-breakdown
NoMoneyDaddy
Break a spec into small verifiable tasks with clear acceptance criteria.
SkillCFreshness: CClaude Code +9 - 0.474score
biome-quality-automation
NoMoneyDaddy
Use Biome as the JavaScript and TypeScript quality loop for formatting, linting, import sorting, safe fixes, and CI checks.
SkillCFreshness: CClaude Code +9 - 0.474score
api-and-interface-design
NoMoneyDaddy
Design stable, explicit, hard-to-misuse interfaces across modules, APIs, services, and data contracts.
SkillCFreshness: CClaude Code +9 - 0.474score
cavekit-backprop
NoMoneyDaddy
Turn bugs and failed tests into durable fixes by updating code, tests, and `docs/SPEC.md`.
SkillCFreshness: CClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get loop-engineering running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
loop-engineering skill. Agent Mart. https://agentmart.dev/skills/nomoneydaddy-loop-engineering-d5a7f3 (last updated 2026-10-24)