@haaaiawd/anws 2.2.2 → 2.2.3

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (36) hide show
  1. package/README.md +180 -180
  2. package/lib/manifest.js +212 -212
  3. package/package.json +1 -1
  4. package/templates/.agents/skills/anws-system/SKILL.md +108 -108
  5. package/templates/.agents/skills/code-reviewer/SKILL.md +101 -101
  6. package/templates/.agents/skills/concept-modeler/SKILL.md +179 -178
  7. package/templates/.agents/skills/craft-authoring/SKILL.md +6 -6
  8. package/templates/.agents/skills/design-reviewer/SKILL.md +190 -176
  9. package/templates/.agents/skills/e2e-testing-guide/SKILL.md +204 -59
  10. package/templates/.agents/skills/nexus-mapper/SKILL.md +321 -306
  11. package/templates/.agents/skills/report-template/SKILL.md +92 -85
  12. package/templates/.agents/skills/runtime-inspector/SKILL.md +12 -12
  13. package/templates/.agents/skills/sequential-thinking/SKILL.md +225 -216
  14. package/templates/.agents/skills/spec-writer/SKILL.md +9 -9
  15. package/templates/.agents/skills/spec-writer/references/prd_template.md +6 -6
  16. package/templates/.agents/skills/system-architect/SKILL.md +678 -620
  17. package/templates/.agents/skills/system-designer/SKILL.md +601 -534
  18. package/templates/.agents/skills/system-designer/references/system-design-detail-template.md +5 -5
  19. package/templates/.agents/skills/system-designer/references/system-design-template.md +28 -28
  20. package/templates/.agents/skills/task-planner/SKILL.md +699 -629
  21. package/templates/.agents/skills/task-planner/references/TASK_TEMPLATE.md +15 -15
  22. package/templates/.agents/skills/task-reviewer/SKILL.md +388 -363
  23. package/templates/.agents/skills/tech-evaluator/SKILL.md +144 -135
  24. package/templates/.agents/skills/tech-evaluator/references/ADR_TEMPLATE.md +80 -78
  25. package/templates/.agents/workflows/blueprint.md +391 -391
  26. package/templates/.agents/workflows/challenge.md +52 -52
  27. package/templates/.agents/workflows/change.md +346 -346
  28. package/templates/.agents/workflows/craft.md +11 -11
  29. package/templates/.agents/workflows/design-system.md +631 -631
  30. package/templates/.agents/workflows/explore.md +399 -399
  31. package/templates/.agents/workflows/forge.md +75 -73
  32. package/templates/.agents/workflows/genesis.md +353 -353
  33. package/templates/.agents/workflows/probe.md +243 -243
  34. package/templates/.agents/workflows/quickstart.md +123 -123
  35. package/templates/.agents/workflows/upgrade.md +10 -10
  36. package/templates/AGENTS.md +149 -149
package/README.md CHANGED
@@ -1,180 +1,180 @@
1
-
2
- <div align="center">
3
-
4
- <img src="assets/logo-cli.png" width="260" alt="Anws">
5
-
6
- [![License: MIT](https://opensource.org/licenses/MIT)](https://opensource.org/licenses/MIT)
7
- [![Version](https://img.shields.io/badge/version-v2.2.2-7FB5B6)](https://github.com/Haaaiawd/ANWS/releases)
8
- [![Targets](https://img.shields.io/badge/Targets-Windsurf%20%7C%20Claude%20Code%20%7C%20Copilot%20%7C%20Cursor%20%7C%20Codex%20Preview%20%7C%20OpenCode%20%7C%20Trae%20%7C%20Qoder%20%7C%20Kilo%20Code-blueviolet)](https://github.com/Haaaiawd/ANWS)
9
-
10
- [English](./README.md) | [中文](./README_CN.md)
11
-
12
- </div>
13
-
14
- ---
15
-
16
- # Anws
17
-
18
- **Anws** is a spec-driven workflow framework for AI-assisted development across modern AI IDEs and coding tools.
19
-
20
- It helps teams build production-ready software through a disciplined path:
21
-
22
- `PRD -> Architecture -> ADR -> Tasks -> Review -> Code -> Upgrade`
23
-
24
- Anws enforces design-first principles, preserves context in files, and prevents architectural drift across multi-tool AI coding workflows.
25
-
26
- > **TL;DR**: a design-first workflow framework for AI coding tools that turns vibe coding into production-oriented engineering.
27
-
28
- ## ANWS
29
-
30
- - **Axiom** — principle before implementation
31
- - **Nexus** — connection before fragmentation
32
- - **Weave** — coherence before accumulation
33
- - **Sovereignty** — human judgment before automation
34
-
35
- ---
36
-
37
- ## Why Anws Exists
38
-
39
- Modern AI coding sessions fail in predictable ways:
40
-
41
- - **Architecture drift**
42
- - different sessions generate incompatible structures
43
- - **Context amnesia**
44
- - a fresh chat loses system decisions, trade-offs, and task state
45
- - **Planning collapse**
46
- - code gets written before requirements and interfaces are stabilized
47
- - **Unsafe upgrades**
48
- - workflow files change over time, but existing projects cannot be updated cleanly
49
-
50
- Anws addresses those problems with:
51
-
52
- - **Versioned architecture docs** under `.anws/v{N}/`
53
- - **A root anchor file** via `AGENTS.md`
54
- - **Workflow-first execution** instead of prompt-only improvisation
55
- - **Controlled update semantics** for `AGENTS.md`, installed targets, and upgrade history
56
-
57
- ---
58
-
59
- ## What's New in v2.2.2
60
-
61
- **v2.2.2** centers the **`/forge`** chain: **`/forge` AUTO** keeps checkpoint ceremony with **`AUTO`** signatures; **`code-reviewer`** follows **wave cadence** (typically **once per wave** at the wave’s last task, with a **~2–3 wave** catch-up if long skipped—not after every task by default); **`e2e-testing-guide`** is **guide first**, then live browser steps when tooling exists—otherwise **guide-only**, never claim “tested” without evidence. **`/change`** handles doc/task backflow only—it **does not run `code-reviewer`** (static fidelity stays in **`/forge` §3.4.5** and **`/challenge`**).
62
-
63
- **v2.2.0** shipped closed-loop **challenge**, explicit contract closure, **`code-reviewer`** as static evidence, forge challenge-report gates, and **`e2e-testing-guide`**. Full history: **[RELEASE_NOTES.md](../RELEASE_NOTES.md)** · [GitHub Releases](https://github.com/Haaaiawd/ANWS/releases).
64
-
65
- **v2.0.0** was the protocol-level major (unified **`.anws/`**, controlled **`AGENTS.md`**, multi-target projection). See **[RELEASE_NOTES.md](../RELEASE_NOTES.md)** for the complete major changelog.
66
-
67
- ---
68
-
69
- ## Quick Start
70
-
71
- ### Install via npm
72
-
73
- ```bash
74
- npm install -g @haaaiawd/anws
75
- cd your-project
76
- anws init
77
- ```
78
-
79
- - **Requirement**
80
- - Node.js `>= 18`
81
- - **Install behavior**
82
- - `anws init` installs one or more target projections into their native folders
83
- - example: `anws init --target windsurf,opencode`
84
-
85
- ### Update an Existing Project
86
-
87
- ```bash
88
- cd your-project
89
- anws update
90
- ```
91
-
92
- - **Update flags**
93
- - `anws update --check` and `anws update --target` are removed; run `anws update` once to refresh all matched targets
94
- - **State source**
95
- - `anws update` reads `.anws/install-lock.json`
96
- - if the lock is missing or invalid, it falls back to directory scan
97
- - if lock drift is detected, directory scan becomes the effective source for the current update
98
- - a real `anws update` can rebuild `.anws/install-lock.json` from detected targets when fallback is active
99
- - **`AGENTS.md` behavior**
100
- - marker-based file -> update stable sections, preserve `AUTO` block
101
- - recognized legacy file -> migrate into new marker-based structure
102
- - unrecognized legacy file -> warn and preserve unchanged
103
- - **Legacy migration**
104
- - if a project still has `.agent/`, the CLI can guide migration to `.agents/`
105
- - after successful migration, interactive mode can also ask whether to delete the old `.agent/`
106
- - **Upgrade record**
107
- - every successful update refreshes `.anws/changelog/`
108
- - target state is written back to `.anws/install-lock.json`
109
-
110
- ---
111
-
112
- ## Feature demos
113
-
114
- What using Anws looks like in practice: architecture-first **`/genesis`**, **human-in-the-loop** requirement alignment, and **skill** orchestration.
115
-
116
- **Deep Thinking & Architecture Design**
117
- <img src="assets/genesis工作流演示.jpg" width="800" alt="Genesis Workflow">
118
-
119
- **Interactive Requirement Alignment**
120
- <img src="assets/与人类交互确认细节.jpg" width="800" alt="Human Interaction">
121
-
122
- **Autonomous Skill Invocation**
123
- <img src="assets/自主调用skills.jpg" width="800" alt="Skills Execution">
124
-
125
- ---
126
-
127
- ## Philosophy
128
-
129
- **1. Docs first—specs keep you in command**
130
- PRD, architecture, tasks, and design land in the repo before code does—so the project doesn’t drift in aimless “vibe runs.” Scope and progress live in `.anws/`, `05_TASKS.md`, and **`AGENTS.md`**: you stay **in control of the system**, not whichever chat window is open.
131
-
132
- **2. Full autonomy inside the rails**
133
- **`/forge` AUTO** is delegation with checkpoints: keep moving inside agreed contracts. **Code review**, **e2e-testing-guide**, and the rest of the template gates keep runs **auditable** and **bounded**. When a wave is executing, it’s reasonable to **walk away—coffee, a walk**—because confidence comes from the spec and gates, not from staring at the model.
134
-
135
- **Iteration is the product**
136
- **`/challenge`** isn’t a one-time rubber stamp; it’s repeated adversarial passes. **Good products and clear ideas are sharpened over cycles**—same as real shipping: each round pulls design, tasks, and implementation back into alignment.
137
-
138
- ---
139
-
140
- ## Recommended Workflow
141
-
142
- Use Anws as a lifecycle, not just a folder pack.
143
-
144
-
145
- | Command | Purpose | Input | Output |
146
- | ----------------- | --------------------------------------------------------------------------- | --------------------- | ------------------------ |
147
- | **`/quickstart`** | Route the user through the correct workflow path | Auto-detected state | Full orchestration |
148
- | `/genesis` | Start from zero with PRD and architecture | Vague idea | PRD, architecture, ADRs |
149
- | `/probe` | Analyze a legacy codebase before change | Existing code | Risk report |
150
- | `/design-system` | Design one system in depth | Architecture overview | System design doc |
151
- | `/challenge` | Review design, tasks, and implementation fidelity with adversarial pressure | Docs / tasks / code | Challenge report |
152
- | `/blueprint` | Break architecture into executable work | PRD + architecture | `05_TASKS.md` |
153
- | `/forge` | Turn approved tasks into code with challenge-report and contract gates | Tasks + review state | Working implementation |
154
- | `/change` | In-version task/contract tweaks (controlled expansion: few new tasks) | Small scoped change | Updated task/design docs |
155
- | `/explore` | Research ambiguous or strategic topics | Topic | Exploration report |
156
- | `/craft` | Create workflows, skills, and prompts | Creation request | Reusable assets |
157
- | `/upgrade` | Route post-update upgrade work | Update changelog | Change or genesis path |
158
-
159
-
160
- ---
161
-
162
- ## Contributing
163
-
164
- Contributions are welcome. Before opening a PR, make sure changes align with the spec-driven workflow and the target projection model.
165
-
166
- ---
167
-
168
- ## License
169
-
170
- [MIT](LICENSE) © 2026
171
-
172
- ---
173
-
174
- <div align="center">
175
-
176
- **Made for architects who code, and AIs who think.**
177
-
178
- *Good architecture isn't written. It's designed.*
179
-
180
- </div>
1
+
2
+ <div align="center">
3
+
4
+ <img src="assets/logo-cli.png" width="260" alt="Anws">
5
+
6
+ [![License: MIT](https://opensource.org/licenses/MIT)](https://opensource.org/licenses/MIT)
7
+ [![Version](https://img.shields.io/badge/version-v2.2.2-7FB5B6)](https://github.com/Haaaiawd/ANWS/releases)
8
+ [![Targets](https://img.shields.io/badge/Targets-Windsurf%20%7C%20Claude%20Code%20%7C%20Copilot%20%7C%20Cursor%20%7C%20Codex%20Preview%20%7C%20OpenCode%20%7C%20Trae%20%7C%20Qoder%20%7C%20Kilo%20Code-blueviolet)](https://github.com/Haaaiawd/ANWS)
9
+
10
+ [English](./README.md) | [中文](./README_CN.md)
11
+
12
+ </div>
13
+
14
+ ---
15
+
16
+ # Anws
17
+
18
+ **Anws** is a spec-driven workflow framework for AI-assisted development across modern AI IDEs and coding tools.
19
+
20
+ It helps teams build production-ready software through a disciplined path:
21
+
22
+ `PRD -> Architecture -> ADR -> Tasks -> Review -> Code -> Upgrade`
23
+
24
+ Anws enforces design-first principles, preserves context in files, and prevents architectural drift across multi-tool AI coding workflows.
25
+
26
+ > **TL;DR**: a design-first workflow framework for AI coding tools that turns vibe coding into production-oriented engineering.
27
+
28
+ ## ANWS
29
+
30
+ - **Axiom** — principle before implementation
31
+ - **Nexus** — connection before fragmentation
32
+ - **Weave** — coherence before accumulation
33
+ - **Sovereignty** — human judgment before automation
34
+
35
+ ---
36
+
37
+ ## Why Anws Exists
38
+
39
+ Modern AI coding sessions fail in predictable ways:
40
+
41
+ - **Architecture drift**
42
+ - different sessions generate incompatible structures
43
+ - **Context amnesia**
44
+ - a fresh chat loses system decisions, trade-offs, and task state
45
+ - **Planning collapse**
46
+ - code gets written before requirements and interfaces are stabilized
47
+ - **Unsafe upgrades**
48
+ - workflow files change over time, but existing projects cannot be updated cleanly
49
+
50
+ Anws addresses those problems with:
51
+
52
+ - **Versioned architecture docs** under `.anws/v{N}/`
53
+ - **A root anchor file** via `AGENTS.md`
54
+ - **Workflow-first execution** instead of prompt-only improvisation
55
+ - **Controlled update semantics** for `AGENTS.md`, installed targets, and upgrade history
56
+
57
+ ---
58
+
59
+ ## What's New in v2.2.2
60
+
61
+ **v2.2.2** centers the **`/forge`** chain: **`/forge` AUTO** keeps checkpoint ceremony with **`AUTO`** signatures; **`code-reviewer`** follows **wave cadence** (typically **once per wave** at the wave’s last task, with a **~2–3 wave** catch-up if long skipped—not after every task by default); **`e2e-testing-guide`** is **guide first**, then live browser steps when tooling exists—otherwise **guide-only**, never claim “tested” without evidence. **`/change`** handles doc/task backflow only—it **does not run `code-reviewer`** (static fidelity stays in **`/forge` §3.4.5** and **`/challenge`**).
62
+
63
+ **v2.2.0** shipped closed-loop **challenge**, explicit contract closure, **`code-reviewer`** as static evidence, forge challenge-report gates, and **`e2e-testing-guide`**. Full history: **[RELEASE_NOTES.md](../RELEASE_NOTES.md)** · [GitHub Releases](https://github.com/Haaaiawd/ANWS/releases).
64
+
65
+ **v2.0.0** was the protocol-level major (unified **`.anws/`**, controlled **`AGENTS.md`**, multi-target projection). See **[RELEASE_NOTES.md](../RELEASE_NOTES.md)** for the complete major changelog.
66
+
67
+ ---
68
+
69
+ ## Quick Start
70
+
71
+ ### Install via npm
72
+
73
+ ```bash
74
+ npm install -g @haaaiawd/anws
75
+ cd your-project
76
+ anws init
77
+ ```
78
+
79
+ - **Requirement**
80
+ - Node.js `>= 18`
81
+ - **Install behavior**
82
+ - `anws init` installs one or more target projections into their native folders
83
+ - example: `anws init --target windsurf,opencode`
84
+
85
+ ### Update an Existing Project
86
+
87
+ ```bash
88
+ cd your-project
89
+ anws update
90
+ ```
91
+
92
+ - **Update flags**
93
+ - `anws update --check` and `anws update --target` are removed; run `anws update` once to refresh all matched targets
94
+ - **State source**
95
+ - `anws update` reads `.anws/install-lock.json`
96
+ - if the lock is missing or invalid, it falls back to directory scan
97
+ - if lock drift is detected, directory scan becomes the effective source for the current update
98
+ - a real `anws update` can rebuild `.anws/install-lock.json` from detected targets when fallback is active
99
+ - **`AGENTS.md` behavior**
100
+ - marker-based file -> update stable sections, preserve `AUTO` block
101
+ - recognized legacy file -> migrate into new marker-based structure
102
+ - unrecognized legacy file -> warn and preserve unchanged
103
+ - **Legacy migration**
104
+ - if a project still has `.agent/`, the CLI can guide migration to `.agents/`
105
+ - after successful migration, interactive mode can also ask whether to delete the old `.agent/`
106
+ - **Upgrade record**
107
+ - every successful update refreshes `.anws/changelog/`
108
+ - target state is written back to `.anws/install-lock.json`
109
+
110
+ ---
111
+
112
+ ## Feature demos
113
+
114
+ What using Anws looks like in practice: architecture-first **`/genesis`**, **human-in-the-loop** requirement alignment, and **skill** orchestration.
115
+
116
+ **Deep Thinking & Architecture Design**
117
+ <img src="assets/genesis工作流演示.jpg" width="800" alt="Genesis Workflow">
118
+
119
+ **Interactive Requirement Alignment**
120
+ <img src="assets/与人类交互确认细节.jpg" width="800" alt="Human Interaction">
121
+
122
+ **Autonomous Skill Invocation**
123
+ <img src="assets/自主调用skills.jpg" width="800" alt="Skills Execution">
124
+
125
+ ---
126
+
127
+ ## Philosophy
128
+
129
+ **1. Docs first—specs keep you in command**
130
+ PRD, architecture, tasks, and design land in the repo before code does—so the project doesn’t drift in aimless “vibe runs.” Scope and progress live in `.anws/`, `05_TASKS.md`, and **`AGENTS.md`**: you stay **in control of the system**, not whichever chat window is open.
131
+
132
+ **2. Full autonomy inside the rails**
133
+ **`/forge` AUTO** is delegation with checkpoints: keep moving inside agreed contracts. **Code review**, **e2e-testing-guide**, and the rest of the template gates keep runs **auditable** and **bounded**. When a wave is executing, it’s reasonable to **walk away—coffee, a walk**—because confidence comes from the spec and gates, not from staring at the model.
134
+
135
+ **Iteration is the product**
136
+ **`/challenge`** isn’t a one-time rubber stamp; it’s repeated adversarial passes. **Good products and clear ideas are sharpened over cycles**—same as real shipping: each round pulls design, tasks, and implementation back into alignment.
137
+
138
+ ---
139
+
140
+ ## Recommended Workflow
141
+
142
+ Use Anws as a lifecycle, not just a folder pack.
143
+
144
+
145
+ | Command | Purpose | Input | Output |
146
+ | ----------------- | --------------------------------------------------------------------------- | --------------------- | ------------------------ |
147
+ | **`/quickstart`** | Route the user through the correct workflow path | Auto-detected state | Full orchestration |
148
+ | `/genesis` | Start from zero with PRD and architecture | Vague idea | PRD, architecture, ADRs |
149
+ | `/probe` | Analyze a legacy codebase before change | Existing code | Risk report |
150
+ | `/design-system` | Design one system in depth | Architecture overview | System design doc |
151
+ | `/challenge` | Review design, tasks, and implementation fidelity with adversarial pressure | Docs / tasks / code | Challenge report |
152
+ | `/blueprint` | Break architecture into executable work | PRD + architecture | `05_TASKS.md` |
153
+ | `/forge` | Turn approved tasks into code with challenge-report and contract gates | Tasks + review state | Working implementation |
154
+ | `/change` | In-version task/contract tweaks (controlled expansion: few new tasks) | Small scoped change | Updated task/design docs |
155
+ | `/explore` | Research ambiguous or strategic topics | Topic | Exploration report |
156
+ | `/craft` | Create workflows, skills, and prompts | Creation request | Reusable assets |
157
+ | `/upgrade` | Route post-update upgrade work | Update changelog | Change or genesis path |
158
+
159
+
160
+ ---
161
+
162
+ ## Contributing
163
+
164
+ Contributions are welcome. Before opening a PR, make sure changes align with the spec-driven workflow and the target projection model.
165
+
166
+ ---
167
+
168
+ ## License
169
+
170
+ [MIT](LICENSE) © 2026
171
+
172
+ ---
173
+
174
+ <div align="center">
175
+
176
+ **Made for architects who code, and AIs who think.**
177
+
178
+ *Good architecture isn't written. It's designed.*
179
+
180
+ </div>