@relipa/ai-flow-kit 0.0.4-beta.2 → 0.0.4

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 (33) hide show
  1. package/custom/skills/investigate-bug/SKILL.md +1 -1
  2. package/custom/skills/validate-ticket/SKILL.md +1 -1
  3. package/custom/templates/laravel.md +15 -15
  4. package/custom/templates/nestjs.md +72 -72
  5. package/custom/templates/nextjs.md +14 -14
  6. package/custom/templates/nodejs-express.md +73 -73
  7. package/custom/templates/python-django.md +71 -71
  8. package/custom/templates/python-fastapi.md +54 -54
  9. package/custom/templates/reactjs.md +492 -492
  10. package/custom/templates/shared/gate-workflow.md +75 -75
  11. package/custom/templates/spring-boot.md +523 -523
  12. package/custom/templates/tools/claude.md +13 -0
  13. package/custom/templates/tools/copilot.md +12 -8
  14. package/custom/templates/tools/cursor.md +12 -8
  15. package/custom/templates/tools/gemini.md +12 -8
  16. package/custom/templates/tools/generic.md +17 -12
  17. package/custom/templates/vue-nuxt.md +14 -14
  18. package/{AIFLOW.md → docs/AIFLOW.md} +2 -2
  19. package/{CHANGELOG.md → docs/CHANGELOG.md} +14 -3
  20. package/{IMPLEMENTATION_SUMMARY.md → docs/IMPLEMENTATION_SUMMARY.md} +21 -39
  21. package/{QUICK_START.md → docs/QUICK_START.md} +10 -10
  22. package/{README.md → docs/README.md} +12 -7
  23. package/docs/ai-integration.md +53 -53
  24. package/docs/architecture.md +5 -6
  25. package/docs/cli-reference.md +2 -2
  26. package/docs/developer-overview.md +126 -126
  27. package/docs/troubleshooting.md +1 -1
  28. package/package.json +2 -8
  29. package/scripts/hooks/session-start.js +2 -2
  30. package/scripts/init.js +501 -460
  31. package/scripts/prompt.js +1 -1
  32. package/scripts/use.js +594 -594
  33. package/CONTRIBUTING.md +0 -388
@@ -0,0 +1,13 @@
1
+ # Claude AI System Instructions
2
+
3
+ You are an expert AI assistant specialized in this project's stack. Follow the Gate Workflow and Team Rules strictly.
4
+
5
+ > **Important:** Always read `.aiflow/context/current.json` to load ticket context and check the `plan/` directory for existing progress before starting any task.
6
+
7
+ If Gate 1 does not auto-start, wait for the developer to type **"start"**, **"Gate 1"** or **"Analyze ticket"**.
8
+
9
+ ## Interaction Rules:
10
+
11
+ - **COLLABORATIVE SKILLS:** When a skill (like `brainstorming`) says to "ask one question at a time", you MUST stop and wait for the developer's reply before proceeding.
12
+ - **NEVER BATCH QUESTIONS:** Only ask one question per message.
13
+ - **WAIT FOR APPROVAL:** Do not move to the next Gate until you receive "APPROVED".
@@ -1,8 +1,12 @@
1
- # GitHub Copilot Custom Instructions
2
-
3
- You are an expert AI assistant specialized in this project's stack. Follow the Gate Workflow and Team Rules strictly.
4
-
5
- > [!IMPORTANT]
6
- > Always check for context in `.claude/context/current.json` and progress in the `plan/` folder before suggesting changes.
7
-
8
- If the Gate Workflow hasn't started, wait for the developer to type **"start"** or **"Gate 1"**.
1
+ # GitHub Copilot Custom Instructions
2
+
3
+ You are an expert AI assistant specialized in this project's stack. Follow the Gate Workflow and Team Rules strictly.
4
+
5
+ > **Important:** Always check for context in `.aiflow/context/current.json` and progress in the `plan/` folder before suggesting changes.
6
+
7
+ If the Gate Workflow hasn't started, wait for the developer to type **"start"** or **"Gate 1"**.
8
+
9
+ ## Interaction Rules:
10
+ - **COLLABORATIVE SKILLS:** When a skill (like `brainstorming`) says to "ask one question at a time", you MUST stop and wait for the developer's reply before proceeding.
11
+ - **NEVER BATCH QUESTIONS:** Only ask one question per message.
12
+ - **WAIT FOR APPROVAL:** Do not move to the next Gate until you receive "APPROVED".
@@ -1,8 +1,12 @@
1
- # Cursor AI Project Rules
2
-
3
- You are an expert AI assistant specialized in this project's stack. Follow the Gate Workflow and Team Rules strictly.
4
-
5
- > [!IMPORTANT]
6
- > Always check `.claude/context/current.json` and the `plan/` directory before starting any task to understand current context and progress.
7
-
8
- If Gate 1 doesn't auto-start, wait for the developer to type **"start"**, **"Gate 1"** or **"Analyze ticket"**.
1
+ # Cursor AI Project Rules
2
+
3
+ You are an expert AI assistant specialized in this project's stack. Follow the Gate Workflow and Team Rules strictly.
4
+
5
+ > **Important:** Always check `.aiflow/context/current.json` and the `plan/` directory before starting any task to understand current context and progress.
6
+
7
+ If Gate 1 doesn't auto-start, wait for the developer to type **"start"**, **"Gate 1"** or **"Analyze ticket"**.
8
+
9
+ ## Interaction Rules:
10
+ - **COLLABORATIVE SKILLS:** When a skill (like `brainstorming`) says to "ask one question at a time", you MUST stop and wait for the developer's reply before proceeding.
11
+ - **NEVER BATCH QUESTIONS:** Only ask one question per message.
12
+ - **WAIT FOR APPROVAL:** Do not move to the next Gate until you receive "APPROVED".
@@ -1,8 +1,12 @@
1
- # Gemini AI System Instructions
2
-
3
- You are an expert AI assistant specialized in this project's stack. Follow the Gate Workflow and Team Rules strictly.
4
-
5
- > [!IMPORTANT]
6
- > When starting a session, always read `.claude/context/current.json` to load ticket context and check the `plan/` directory for existing progress.
7
-
8
- If no instructions are automatically followed, wait for the developer to type **"start"** or **"Gate 1"** to start the analysis.
1
+ # Gemini AI System Instructions
2
+
3
+ You are an expert AI assistant specialized in this project's stack. Follow the Gate Workflow and Team Rules strictly.
4
+
5
+ > **Important:** When starting a session, always read `.aiflow/context/current.json` to load ticket context and check the `plan/` directory for existing progress.
6
+
7
+ If no instructions are automatically followed, wait for the developer to type **"start"** or **"Gate 1"** to start the analysis.
8
+
9
+ ## Interaction Rules:
10
+ - **COLLABORATIVE SKILLS:** When a skill (like `brainstorming`) says to "ask one question at a time", you MUST stop and wait for the developer's reply before proceeding.
11
+ - **NEVER BATCH QUESTIONS:** Only ask one question per message.
12
+ - **WAIT FOR APPROVAL:** Do not move to the next Gate until you receive "APPROVED".
@@ -1,12 +1,17 @@
1
- # AI System Instructions
2
-
3
- You are an expert AI assistant. Follow the project's Gate Workflow and Team Rules strictly.
4
-
5
- ## Context Awareness
6
- - **Ticket context**: `.claude/context/current.json`
7
- - **Task progress**: `plan/[ticket-id]/`
8
- - **Rules**: `.rules/`
9
-
10
- Follow the gated workflow and rules defined below.
11
-
12
- If Gate 1 does not start automatically, please start it when the developer types **"Gate 1"**.
1
+ # AI System Instructions
2
+
3
+ You are an expert AI assistant. Follow the project's Gate Workflow and Team Rules strictly.
4
+
5
+ ## Context Awareness
6
+ - **Ticket context**: `.aiflow/context/current.json`
7
+ - **Task progress**: `plan/[ticket-id]/`
8
+ - **Rules**: `.rules/`
9
+
10
+ Follow the gated workflow and rules defined below.
11
+
12
+ If Gate 1 does not start automatically, please start it when the developer types **"Gate 1"**.
13
+
14
+ ## Interaction Rules:
15
+ - **COLLABORATIVE SKILLS:** When a skill (like `brainstorming`) says to "ask one question at a time", you MUST stop and wait for the developer's reply before proceeding.
16
+ - **NEVER BATCH QUESTIONS:** Only ask one question per message.
17
+ - **WAIT FOR APPROVAL:** Do not move to the next Gate until you receive "APPROVED".
@@ -1,14 +1,14 @@
1
- # Vue/Nuxt.js AI System Prompt
2
-
3
- You are an expert Vue.js & Nuxt 3 developer. Follow these rules.
4
-
5
- ---
6
-
7
- - Use `<script setup lang="ts">` with Composition API exclusively. Do not use the Options API.
8
- - Use explicit Nuxt auto-imports where appropriate, but maintain clarity on complex utilities.
9
- - For state management, prefer Pinia or native `useState` bindings in Nuxt.
10
- - Use `useFetch` or `useAsyncData` for server-side data fetching. Use `$fetch` for client-side API requests outside component initialization.
11
- - Keep components small and specialized.
12
- - Adhere to the official Vue Style Guide (Priority A and B).
13
-
14
- Ensure code is strictly typed with TypeScript.
1
+ # Vue/Nuxt.js AI System Prompt
2
+
3
+ You are an expert Vue.js & Nuxt 3 developer. Follow these rules.
4
+
5
+ ---
6
+
7
+ - Use `<script setup lang="ts">` with Composition API exclusively. Do not use the Options API.
8
+ - Use explicit Nuxt auto-imports where appropriate, but maintain clarity on complex utilities.
9
+ - For state management, prefer Pinia or native `useState` bindings in Nuxt.
10
+ - Use `useFetch` or `useAsyncData` for server-side data fetching. Use `$fetch` for client-side API requests outside component initialization.
11
+ - Keep components small and specialized.
12
+ - Adhere to the official Vue Style Guide (Priority A and B).
13
+
14
+ Ensure code is strictly typed with TypeScript.
@@ -83,7 +83,7 @@ claude # open Claude → AI auto-starts Gate 1
83
83
 
84
84
  | Step | Action | Purpose |
85
85
  |------|--------|---------|
86
- | 1 | Load ticket from `.claude/context/current.json` | Get PM's requirements, comments |
86
+ | 1 | Load ticket from `.aiflow/context/current.json` | Get PM's requirements, comments |
87
87
  | 2 | Read `CLAUDE.md` + source code | Understand architecture, tech stack, patterns |
88
88
  | 3 | Investigate related files & data flow | Identify affected areas, dependencies |
89
89
  | 4 | Ask clarifying questions (one at a time) | Ensure full understanding |
@@ -415,7 +415,7 @@ aiflow guide # view quickstart guide
415
415
 
416
416
  ## Multi-AI Environment Setup
417
417
 
418
- One of the core strengths of `ai-flow-kit` is that the **Gate Workflow** is tool-agnostic. Since state is saved in `.claude/context/current.json` and the `plan/` directory, you can switch tools mid-task.
418
+ One of the core strengths of `ai-flow-kit` is that the **Gate Workflow** is tool-agnostic. Since state is saved in `.aiflow/context/current.json` and the `plan/` directory, you can switch tools mid-task.
419
419
 
420
420
  ### How to Switch Tools
421
421
  1. **Analyze (Gate 1)** using Claude Code CLI (great for deep codebase scans).
@@ -7,9 +7,21 @@ Versioning follows [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
7
7
 
8
8
  ---
9
9
 
10
- ## [Unreleased]
10
+ ## [0.0.4-beta.5] - 2026-04-21
11
+
12
+ ### Added
13
+ - Created a dedicated `docs/` folder to house all developer-facing documentation.
14
+ - Integrated AI Skill Registry into all tool templates (Claude, Cursor, Gemini, Copilot) for better skill discovery.
15
+
16
+ ### Changed
17
+ - Moved `README.md`, `QUICK_START.md`, `AIFLOW.md`, `CHANGELOG.md`, and `IMPLEMENTATION_SUMMARY.md` into the `docs/` directory.
18
+ - Updated `package.json` to exclude internal-only files (`CONTRIBUTING.md`, `plan.md`) from the NPM package distribution.
19
+ - Updated `scripts/init.js` to source documentation from the new `docs/` location.
20
+ - Updated all internal documentation links to reflect the new directory structure.
21
+
22
+ ---
11
23
 
12
- > Changes that are merged but not yet released.
24
+ ## [Unreleased]
13
25
 
14
26
  ---
15
27
 
@@ -64,7 +76,6 @@ Versioning follows [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
64
76
  #### Project Templates
65
77
  - `AIFLOW.md` — team workflow reference document
66
78
  - `QUICK_START.md` — 5-minute setup guide
67
- - `CONTRIBUTING.md` — contributor guide
68
79
  - `.aiflowrc.json.example` — configuration file reference
69
80
 
70
81
  ### Architecture Notes
@@ -64,7 +64,7 @@ Complete overview of what has been implemented.
64
64
  ### Tier 2: Developer Experience (COMPLETED)
65
65
 
66
66
  #### 5. ✅ Documentation & Onboarding
67
- - **Files:** `docs/*.md`, `README.md`, `CONTRIBUTING.md`, `QUICK_START.md`
67
+ - **Files:** `docs/*.md`
68
68
  - **Documentation Includes:**
69
69
  - **README.md** — Package overview, features, usage
70
70
  - **QUICK_START.md** — 5-minute getting started
@@ -73,7 +73,6 @@ Complete overview of what has been implemented.
73
73
  - **docs/configuration.md** — All configuration options
74
74
  - **docs/cli-reference.md** — Complete CLI command reference
75
75
  - **docs/troubleshooting.md** — Common issues and solutions
76
- - **CONTRIBUTING.md** — How to extend the package
77
76
  - **Workflow Guides:**
78
77
  - `docs/workflows/bug-fix.md` — Fix production issues
79
78
  - `docs/workflows/feature.md` — Build new features
@@ -185,6 +184,23 @@ Complete overview of what has been implemented.
185
184
  ai-flow-kit/
186
185
  ├── bin/
187
186
  │ └── aiflow.js # CLI entry point
187
+ ├── docs/ # Documentation and Guides
188
+ │ ├── README.md # Package overview
189
+ │ ├── QUICK_START.md # Quick start guide
190
+ │ ├── AIFLOW.md # Workflow (5 Gates) detail
191
+ │ ├── CHANGELOG.md # Version history
192
+ │ ├── IMPLEMENTATION_SUMMARY.md # Technical implementation details
193
+ │ ├── getting-started.md
194
+ │ ├── architecture.md
195
+ │ ├── configuration.md
196
+ │ ├── cli-reference.md
197
+ │ ├── troubleshooting.md
198
+ │ └── workflows/
199
+ │ ├── bug-fix.md
200
+ │ ├── feature.md
201
+ │ ├── investigation.md
202
+ │ ├── refactor.md
203
+ │ └── impact-analysis.md
188
204
  ├── scripts/
189
205
  │ ├── init.js # Project initialization
190
206
  │ ├── update.js # Version management
@@ -196,50 +212,16 @@ ai-flow-kit/
196
212
  │ └── memory.js # Knowledge management
197
213
  ├── custom/
198
214
  │ ├── skills/ # Team-specific skills
199
- │ │ ├── investigate-bug/
200
- │ │ ├── impact-analysis/
201
- │ │ ├── generate-spec/
202
- │ │ └── report-customer/
203
215
  │ ├── rules/ # Team standards
204
- ├── code-style.md
205
- │ │ ├── naming.md
206
- │ │ └── review-checklist.md
207
- │ ├── templates/ # Framework rules
208
- │ │ ├── laravel.md
209
- │ │ ├── nextjs.md
210
- │ │ └── vue-nuxt.md
216
+ │ ├── templates/ # Tool templates
211
217
  │ ├── prompts/ # Task templates
212
- │ │ ├── bug-fix.md
213
- │ │ ├── feature.md
214
- │ │ └── investigation.md
215
218
  │ └── mcp-presets/ # MCP adapters
216
- │ ├── jira.json
217
- │ ├── backlog.json
218
- │ ├── google-sheets.json
219
- │ └── README.md
220
219
  ├── upstream/ # Original obra/superpowers
221
- │ ├── skills/
222
- │ ├── agents/
223
- │ └── commands/
224
- ├── docs/
225
- │ ├── getting-started.md
226
- │ ├── architecture.md
227
- │ ├── configuration.md
228
- │ ├── cli-reference.md
229
- │ ├── troubleshooting.md
230
- │ └── workflows/
231
- │ ├── bug-fix.md
232
- │ ├── feature.md
233
- │ ├── investigation.md
234
- │ ├── refactor.md
235
- │ └── impact-analysis.md
236
220
  ├── package.json # Updated with metadata
237
221
  ├── .gitignore # Updated
238
222
  ├── .aiflowrc.json.example # Configuration template
239
- ├── README.md # Complete documentation
240
- ├── QUICK_START.md # Quick getting started
241
- ├── CONTRIBUTING.md # Contribution guide
242
- └── index.js # Main entry point
223
+ ├── index.js # Main entry point
224
+ └── plan.md # TechLead planning (root)
243
225
  ```
244
226
 
245
227
  ## 🎯 Key Features Summary
@@ -1,6 +1,6 @@
1
1
  # Quick Start — AI Flow Kit
2
2
 
3
- > Full workflow: [AIFLOW.md](./AIFLOW.md) | Contributing: [CONTRIBUTING.md](./CONTRIBUTING.md)
3
+ > Full workflow: [AIFLOW.md](./AIFLOW.md)
4
4
 
5
5
  ---
6
6
 
@@ -103,7 +103,7 @@ claude # AI auto-starts Gate 1 immediately
103
103
  > **Multi-Tool Support:**
104
104
  > - If using **Cursor**: Gate Workflow is enforced via `.cursorrules`.
105
105
  > - If using **Gemini CLI**: Rules are loaded from `GEMINI.md`.
106
- > - The AI will automatically detect if you are mid-task by reading `.claude/context/current.json`.
106
+ > - The AI will automatically detect if you are mid-task by reading `.aiflow/context/current.json`.
107
107
 
108
108
  ### Step 3: GATE 1 — AI Analyzes Requirement
109
109
 
@@ -111,7 +111,7 @@ claude # AI auto-starts Gate 1 immediately
111
111
 
112
112
  When you open Claude, AI will **automatically** start analyzing:
113
113
 
114
- 1. **Read ticket** from `.claude/context/current.json`
114
+ 1. **Read ticket** from `.aiflow/context/current.json`
115
115
  2. **Read source code** — understand architecture, related files, data flow
116
116
  3. **Ask clarifying questions** (one at a time) if anything is unclear
117
117
  4. **Output** `plan/PROJ-33/requirement.md` containing:
@@ -348,14 +348,14 @@ After installing `ai-flow-kit`, these docs are available:
348
348
  | [QUICK_START.md](./QUICK_START.md) | This file — step-by-step guide for developers |
349
349
  | [AIFLOW.md](./AIFLOW.md) | Complete 5-Gate workflow: who does what, outputs, templates |
350
350
  | [README.md](./README.md) | Package overview, architecture, all features |
351
- | [CONTRIBUTING.md](./CONTRIBUTING.md) | How to add skills, templates, adapters, rules |
351
+ | [CHANGELOG.md](./CHANGELOG.md) | Version history |
352
352
  | [IMPLEMENTATION_SUMMARY.md](./IMPLEMENTATION_SUMMARY.md) | Technical implementation details |
353
- | [docs/workflows/](./docs/workflows/) | Per-task-type workflow guides |
354
- | [docs/architecture.md](./docs/architecture.md) | System architecture |
355
- | [docs/cli-reference.md](./docs/cli-reference.md) | CLI command reference |
356
- | [docs/configuration.md](./docs/configuration.md) | Configuration guide |
357
- | [docs/troubleshooting.md](./docs/troubleshooting.md) | Common issues & fixes |
358
- | [docs/getting-started.md](./docs/getting-started.md) | Getting started guide |
353
+ | [workflows/](./workflows/) | Per-task-type workflow guides |
354
+ | [architecture.md](./architecture.md) | System architecture |
355
+ | [cli-reference.md](./cli-reference.md) | CLI command reference |
356
+ | [configuration.md](./configuration.md) | Configuration guide |
357
+ | [troubleshooting.md](./troubleshooting.md) | Common issues & fixes |
358
+ | [getting-started.md](./getting-started.md) | Getting started guide |
359
359
 
360
360
  Access docs after install:
361
361
  ```bash
@@ -25,10 +25,9 @@ Developers only need a single command to load ticket context → AI automaticall
25
25
 
26
26
  - **[Quick Start Guide](./QUICK_START.md)** — Step-by-step instructions for developers.
27
27
  - **[Full Workflow (5 Gates)](./AIFLOW.md)** — In-depth look at the Gate process.
28
- - **[Developer Onboarding](./docs/developer-overview.md)** — Presentation-ready guide for new team members.
29
- - **[Contributing Guide](./CONTRIBUTING.md)** — How to extend with new skills/templates.
30
- - **[Troubleshooting](./docs/troubleshooting.md)** — Common issues and fixes.
31
- - **[Integration Guide](./docs/ai-integration.md)** — Advanced tool setup (Claude, Cursor, Gemini).
28
+ - **[Developer Onboarding](./developer-overview.md)** — Presentation-ready guide for new team members.
29
+ - **[Troubleshooting](./troubleshooting.md)** — Common issues and fixes.
30
+ - **[Integration Guide](./ai-integration.md)** — Advanced tool setup (Claude, Cursor, Gemini).
32
31
 
33
32
  ---
34
33
 
@@ -154,7 +153,7 @@ aiflow init --framework spring-boot,reactjs --adapter backlog,jira
154
153
 
155
154
  ### `aiflow use <ticket>`
156
155
 
157
- Load ticket context into `.claude/context/current.json`.
156
+ Load ticket context into `.aiflow/context/current.json`.
158
157
 
159
158
  ```bash
160
159
  # Backlog
@@ -282,6 +281,12 @@ aiflow memory --list
282
281
  ```
283
282
  ai-flow-kit/
284
283
  ├── bin/aiflow.js # CLI entry
284
+ ├── docs/ # Documentation and Guides
285
+ │ ├── README.md # Main documentation
286
+ │ ├── QUICK_START.md # Quick start guide
287
+ │ ├── AIFLOW.md # Workflow (5 Gates) detail
288
+ │ ├── CHANGELOG.md # Version history
289
+ │ └── ... # Other guides
285
290
  ├── scripts/
286
291
  │ ├── init.js # Project initialization
287
292
  │ ├── update.js # Version management
@@ -297,7 +302,7 @@ ai-flow-kit/
297
302
  │ ├── rules/java/ # Java/Spring Boot rules
298
303
  │ ├── rules/php/ # PHP/Laravel rules
299
304
  │ ├── rules/javascript/ # JS/TS rules
300
- │ ├── templates/ # Framework CLAUDE.md templates
305
+ │ ├── templates/ # Framework tool templates
301
306
  │ ├── prompts/ # Task prompt templates
302
307
  │ └── mcp-presets/ # Backlog/Jira/Sheets config
303
308
  └── upstream/ # obra/superpowers (inherited)
@@ -347,4 +352,4 @@ aiflow init --adapter backlog
347
352
  aiflow doctor
348
353
  ```
349
354
 
350
- See more: [docs/troubleshooting.md](./docs/troubleshooting.md)
355
+ See more: [docs/troubleshooting.md](./troubleshooting.md)
@@ -1,53 +1,53 @@
1
- # Multi-AI Environment Integration
2
-
3
- `ai-flow-kit` is designed to be tool-agnostic. While it works best with Claude Code CLI due to deep MCP integration, you can use it with any AI tool.
4
-
5
- ## Supported Environments
6
-
7
- | Tool | Instruction File | Support Level | Key Features |
8
- |------|------------------|---------------|--------------|
9
- | **Claude Code** | `CLAUDE.md` | Native (Best) | Auto-triggers, full MCP access, session hooks. |
10
- | **Cursor** | `.cursorrules` | High | Project-wide rules, `@Codebase` context. |
11
- | **Gemini CLI** | `GEMINI.md` | Medium | System instructions support. |
12
- | **GitHub Copilot** | `.github/copilot-instructions.md` | Medium | Inline suggestions, custom instructions. |
13
- | **OpenCode / Generic** | `AI_INSTRUCTIONS.md` | General | Manual context loading. |
14
-
15
- ---
16
-
17
- ## Tool-Specific Setup
18
-
19
- ### 1. Cursor AI
20
- When you run `aiflow init`, it generates a `.cursorrules` file.
21
- - Cursor will automatically read this file to understand the **Gate Workflow**.
22
- - Use `@Codebase` when you need the AI to search across your whole project.
23
- - **Resumption**: If you start a task in Claude Code and switch to Cursor, Cursor will see the generated `plan/` files and resume from where you left off.
24
-
25
- ### 2. Gemini CLI
26
- Pass `GEMINI.md` as system instructions to your Gemini session.
27
- - **Tip**: Always ensure the AI reads `.claude/context/current.json` first to load the ticket context.
28
- - Gemini is excellent for large context analysis during Gate 1.
29
-
30
- ### 3. GitHub Copilot / Codex
31
- Copilot uses `.github/copilot-instructions.md` to guide its suggestions.
32
- - This ensures that Copilot's inline completions follow your project's architecture and team rules.
33
-
34
- ---
35
-
36
- ## Cross-Tool State Resumption
37
- The "Secret Sauce" of `ai-flow-kit` is its file-based state management:
38
-
39
- 1. **Context**: Saved in `.claude/context/current.json`.
40
- 2. **Progress**: Saved as Markdown docs in `plan/[ticket-id]/`.
41
- 3. **Rules**: Saved in `.rules/`.
42
-
43
- Because these are standard files, any AI tool can read them. You can:
44
- 1. Start **Gate 1** (Analysis) in Claude Code.
45
- 2. Review and **Approve** the requirement doc.
46
- 3. Switch to **Cursor** for **Gate 3** (Coding) because you prefer the IDE integrations.
47
- 4. Switch back to **Claude Code** for **Gate 4** (Self-Review) to use its automated testing power.
48
-
49
- ---
50
-
51
- ## Best Practices
52
- - **Always Approve Docs**: Ensure `requirement.md` and `plan.md` are approved (you can just type "APPROVED" in the chat or edit the file to say so).
53
- - **Update Status**: If you make progress in a tool that doesn't have hooks, manually update the `summary.md` or the ticket status.
1
+ # Multi-AI Environment Integration
2
+
3
+ `ai-flow-kit` is designed to be tool-agnostic. While it works best with Claude Code CLI due to deep MCP integration, you can use it with any AI tool.
4
+
5
+ ## Supported Environments
6
+
7
+ | Tool | Instruction File | Support Level | Key Features |
8
+ |------|------------------|---------------|--------------|
9
+ | **Claude Code** | `CLAUDE.md` | Native (Best) | Auto-triggers, full MCP access, session hooks. |
10
+ | **Cursor** | `.cursorrules` | High | Project-wide rules, `@Codebase` context. |
11
+ | **Gemini CLI** | `GEMINI.md` | Medium | System instructions support. |
12
+ | **GitHub Copilot** | `.github/copilot-instructions.md` | Medium | Inline suggestions, custom instructions. |
13
+ | **OpenCode / Generic** | `AI_INSTRUCTIONS.md` | General | Manual context loading. |
14
+
15
+ ---
16
+
17
+ ## Tool-Specific Setup
18
+
19
+ ### 1. Cursor AI
20
+ When you run `aiflow init`, it generates a `.cursorrules` file.
21
+ - Cursor will automatically read this file to understand the **Gate Workflow**.
22
+ - Use `@Codebase` when you need the AI to search across your whole project.
23
+ - **Resumption**: If you start a task in Claude Code and switch to Cursor, Cursor will see the generated `plan/` files and resume from where you left off.
24
+
25
+ ### 2. Gemini CLI
26
+ Pass `GEMINI.md` as system instructions to your Gemini session.
27
+ - **Tip**: Always ensure the AI reads `.aiflow/context/current.json` first to load the ticket context.
28
+ - Gemini is excellent for large context analysis during Gate 1.
29
+
30
+ ### 3. GitHub Copilot / Codex
31
+ Copilot uses `.github/copilot-instructions.md` to guide its suggestions.
32
+ - This ensures that Copilot's inline completions follow your project's architecture and team rules.
33
+
34
+ ---
35
+
36
+ ## Cross-Tool State Resumption
37
+ The "Secret Sauce" of `ai-flow-kit` is its file-based state management:
38
+
39
+ 1. **Context**: Saved in `.aiflow/context/current.json`.
40
+ 2. **Progress**: Saved as Markdown docs in `plan/[ticket-id]/`.
41
+ 3. **Rules**: Saved in `.rules/`.
42
+
43
+ Because these are standard files, any AI tool can read them. You can:
44
+ 1. Start **Gate 1** (Analysis) in Claude Code.
45
+ 2. Review and **Approve** the requirement doc.
46
+ 3. Switch to **Cursor** for **Gate 3** (Coding) because you prefer the IDE integrations.
47
+ 4. Switch back to **Claude Code** for **Gate 4** (Self-Review) to use its automated testing power.
48
+
49
+ ---
50
+
51
+ ## Best Practices
52
+ - **Always Approve Docs**: Ensure `requirement.md` and `plan.md` are approved (you can just type "APPROVED" in the chat or edit the file to say so).
53
+ - **Update Status**: If you make progress in a tool that doesn't have hooks, manually update the `summary.md` or the ticket status.
@@ -233,7 +233,7 @@ aiflow use JIRA-123
233
233
  [Load from MCP Adapter]
234
234
  - Fetch issue from Jira
235
235
  - Parse description, criteria
236
- - Save to .claude/context/current.json
236
+ - Save to .aiflow/context/current.json
237
237
 
238
238
  aiflow prompt bug-fix
239
239
 
@@ -329,10 +329,9 @@ ai-flow-kit/
329
329
  │ │ └── refactor.md
330
330
  │ └── troubleshooting.md
331
331
 
332
+ ├── docs/ # Documentation
332
333
  ├── package.json
333
- ├── README.md
334
- ├── CONTRIBUTING.md
335
- └── .gitignore
334
+ └── index.js
336
335
  ```
337
336
 
338
337
  ## Customization Points
@@ -340,7 +339,7 @@ ai-flow-kit/
340
339
  ### Add a Custom Skill
341
340
 
342
341
  1. Create `custom/skills/my-skill/SKILL.md`
343
- 2. Follow [skill format](../CONTRIBUTING.md#1-add-a-custom-skill)
342
+ 2. Follow skill format (see technical documentation)
344
343
  3. Run `aiflow update` to activate
345
344
  4. Skill appears in `.claude/skills/`
346
345
 
@@ -374,7 +373,7 @@ The system is designed to extend:
374
373
  - **Validators:** Check output quality
375
374
  - **Plugins:** Custom commands and functionality
376
375
 
377
- See [Contributing Guide](../CONTRIBUTING.md) for details.
376
+ See technical documentation for details.
378
377
 
379
378
  ## Best Practices
380
379
 
@@ -80,7 +80,7 @@ aiflow use 1.0.0
80
80
  **What it does:**
81
81
  1. Fetches context from MCP adapter (if applicable)
82
82
  2. Parses and validates context
83
- 3. Saves to `.claude/context/current.json`
83
+ 3. Saves to `.aiflow/context/current.json`
84
84
  4. Updates state tracking
85
85
 
86
86
  **Example:**
@@ -90,7 +90,7 @@ Fetching context from Jira...
90
90
  ✓ Title: Fix payment processing error
91
91
  ✓ Description: 2000+ words pulled
92
92
  ✓ Acceptance criteria: 3 items
93
- ✓ Saved to .claude/context/current.json
93
+ ✓ Saved to .aiflow/context/current.json
94
94
 
95
95
  Ready to prompt! Run: aiflow prompt bug-fix
96
96
  ```