@relipa/ai-flow-kit 0.0.5-beta.0 → 0.0.5

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 (35) hide show
  1. package/README.md +23 -17
  2. package/custom/skills/{validate-ticket → read-study-requirement}/SKILL.md +27 -17
  3. package/custom/skills/review-plan/SKILL.md +1 -1
  4. package/custom/templates/shared/gate-workflow.md +88 -88
  5. package/custom/templates/tools/claude.md +1 -1
  6. package/custom/templates/tools/copilot.md +1 -1
  7. package/custom/templates/tools/cursor.md +1 -1
  8. package/custom/templates/tools/gemini.md +1 -1
  9. package/custom/templates/tools/generic.md +1 -1
  10. package/docs/{AIFLOW.md → common/AIFLOW.md} +462 -463
  11. package/docs/{CHANGELOG.md → common/CHANGELOG.md} +132 -128
  12. package/docs/{cli-reference.md → common/cli-reference.md} +1 -1
  13. package/docs/project/ARCHITECTURE.md +28 -0
  14. package/package.json +3 -2
  15. package/scripts/context.js +1 -1
  16. package/scripts/hooks/session-start.js +145 -145
  17. package/scripts/init.js +154 -47
  18. package/scripts/prompt.js +431 -402
  19. package/scripts/telemetry/cli.js +6 -12
  20. package/scripts/telemetry/config.js +1 -4
  21. package/scripts/telemetry/flush.js +16 -13
  22. package/scripts/use.js +74 -31
  23. package/docs/IMPLEMENTATION_SUMMARY.md +0 -330
  24. package/docs/architecture.md +0 -394
  25. package/docs/developer-overview.md +0 -126
  26. /package/docs/{QUICK_START.md → common/QUICK_START.md} +0 -0
  27. /package/docs/{ai-integration.md → common/ai-integration.md} +0 -0
  28. /package/docs/{configuration.md → common/configuration.md} +0 -0
  29. /package/docs/{getting-started.md → common/getting-started.md} +0 -0
  30. /package/docs/{troubleshooting.md → common/troubleshooting.md} +0 -0
  31. /package/docs/{workflows → common/workflows}/bug-fix.md +0 -0
  32. /package/docs/{workflows → common/workflows}/feature.md +0 -0
  33. /package/docs/{workflows → common/workflows}/impact-analysis.md +0 -0
  34. /package/docs/{workflows → common/workflows}/investigation.md +0 -0
  35. /package/docs/{workflows → common/workflows}/refactor.md +0 -0
package/README.md CHANGED
@@ -21,13 +21,13 @@ Developers only need a single command to load ticket context → AI automaticall
21
21
  ## Documentation & Guides
22
22
 
23
23
  > [!TIP]
24
- > After running `aiflow init`, all documentation files are automatically copied to your project root and `.aiflow/docs/` for easy access.
24
+ > After running `aiflow init`, documentation files are automatically copied to `.aiflow/docs/` in your project for easy access.
25
25
 
26
- - **[Quick Start Guide](./docs/QUICK_START.md)** — Step-by-step instructions for developers.
27
- - **[Full Workflow (5 Gates)](./docs/AIFLOW.md)** — In-depth look at the Gate process.
28
- - **[Developer Onboarding](./docs/developer-overview.md)** — Presentation-ready guide for new team members.
29
- - **[Troubleshooting](./docs/troubleshooting.md)** — Common issues and fixes.
30
- - **[Integration Guide](./docs/ai-integration.md)** — Advanced tool setup (Claude, Cursor, Gemini).
26
+ - **[Quick Start Guide](./docs/common/QUICK_START.md)** — Step-by-step instructions for developers.
27
+ - **[Full Workflow (5 Gates)](./docs/common/AIFLOW.md)** — In-depth look at the Gate process.
28
+ - **[Troubleshooting](./docs/common/troubleshooting.md)** — Common issues and fixes.
29
+ - **[Integration Guide](./docs/common/ai-integration.md)** — Advanced tool setup (Claude, Cursor, Gemini).
30
+ - **[Changelog](./docs/common/CHANGELOG.md)** — Version history and release notes.
31
31
 
32
32
  ---
33
33
 
@@ -50,8 +50,10 @@ Developers only need a single command to load ticket context → AI automaticall
50
50
  │ .claude/ │
51
51
  │ ├── skills/ ← superpowers + custom│
52
52
  │ ├── hooks/ ← SessionStart hook │
53
- ├── settings.json ← hook config │
54
- └── context/ ← ticket context
53
+ └── settings.json ← hook config │
54
+ .aiflow/
55
+ │ ├── context/ ← ticket context │
56
+ │ └── versions/ ← versioned skills │
55
57
  │ .rules/ ← team rules (java/) │
56
58
  │ CLAUDE.md ← framework template │
57
59
  │ .mcp.json ← MCP adapter config │
@@ -229,9 +231,9 @@ aiflow validate src/Payment.java --ruleset strict
229
231
  Save/search team knowledge.
230
232
 
231
233
  ```bash
232
- aiflow memory --save "payment-flow" "Payment flow description..."
233
- aiflow memory --search "payment"
234
- aiflow memory --list
234
+ aiflow memory save "payment-flow" "Payment flow description..."
235
+ aiflow memory search "payment"
236
+ aiflow memory list
235
237
  ```
236
238
 
237
239
  ---
@@ -283,11 +285,15 @@ aiflow memory --list
283
285
  ai-flow-kit/
284
286
  ├── README.md # Main documentation
285
287
  ├── bin/aiflow.js # CLI entry
286
- ├── docs/ # Documentation and Guides
287
- │ ├── QUICK_START.md # Quick start guide
288
- │ ├── AIFLOW.md # Workflow (5 Gates) detail
289
- │ ├── CHANGELOG.md # Version history
290
- └── ... # Other guides
288
+ ├── docs/
289
+ │ ├── common/ # Published with package
290
+ ├── QUICK_START.md # Quick start guide
291
+ ├── AIFLOW.md # Workflow (5 Gates) detail
292
+ │ ├── CHANGELOG.md # Version history
293
+ │ │ ├── cli-reference.md # Full CLI reference
294
+ │ │ └── workflows/ # Per-task workflow guides
295
+ │ └── project/ # Project template files
296
+ │ └── ARCHITECTURE.md # Architecture doc template
291
297
  ├── scripts/
292
298
  │ ├── init.js # Project initialization
293
299
  │ ├── update.js # Version management
@@ -353,4 +359,4 @@ aiflow init --adapter backlog
353
359
  aiflow doctor
354
360
  ```
355
361
 
356
- See more: [docs/troubleshooting.md](./docs/troubleshooting.md)
362
+ See more: [docs/common/troubleshooting.md](./docs/common/troubleshooting.md)
@@ -1,7 +1,7 @@
1
1
  ---
2
- name: validate-ticket
3
- description: Gate 1 — AI reads ticket from Backlog/Jira + reads source code, then uses superpowers:brainstorming for collaborative Q&A and solution proposal. Adds impact analysis and effort estimate. Outputs requirement.md for DEV to approve.
4
- keywords: ticket, validate, context, requirement, backlog, jira, analyze, understand, solution, estimate
2
+ name: read-study-requirement
3
+ description: Gate 1 — AI reads ticket from Backlog/Jira + reads source code, then uses collaborative Q&A and solution proposal. Adds impact analysis and effort estimate. Outputs requirement.md for DEV to approve.
4
+ keywords: ticket, requirement, study, read, context, backlog, jira, analyze, understand, solution, estimate
5
5
  ---
6
6
 
7
7
  # Analyze Requirement — Gate 1
@@ -19,31 +19,40 @@ keywords: ticket, validate, context, requirement, backlog, jira, analyze, unders
19
19
  1. Read `.aiflow/context/current.json` — ticket info from Backlog/Jira
20
20
  2. Read `CLAUDE.md` — understand project architecture, tech stack, conventions
21
21
  3. Read related source files — trace data flow, identify patterns, dependencies
22
+ - **If task is a Bug Fix:** Invoke `superpowers:systematic-debugging` to trace root cause before proposing fixes.
22
23
  4. Read ticket comments — additional context from PM/team
23
24
 
24
25
  Do this investigation **before** asking any questions. Come prepared.
25
26
 
26
27
  ---
27
28
 
28
- ### Step 2: Clarify Requirements via Brainstorming
29
+ ### Step 2: Clarify Requirements via Q&A
29
30
 
30
- **INVOKE: `superpowers:brainstorming`**
31
+ Ask clarifying questions directly — **one question at a time**, wait for the developer's response before asking the next.
31
32
 
32
- Use brainstorming's collaborative Q&A process to:
33
- - Ask clarifying questions **one at a time** — never batch multiple questions
34
- - Understand purpose, constraints, and success criteria from the developer
35
- - Propose **2–3 solution approaches** with trade-offs and a recommendation
36
- - Present the proposed design and get developer confirmation
33
+ **When to ask:**
34
+ - Business requirements are vague or ambiguous
35
+ - Acceptance criteria are not measurable
36
+ - Edge cases discovered in source code need input
37
+ - Multiple approaches exist and the developer's preference matters
37
38
 
38
- Brainstorming handles the full Q&A loop automatically. Let it run until the developer has confirmed the design direction.
39
+ **When NOT to ask:**
40
+ - The ticket and source code together make the intent clear → proceed directly
39
41
 
40
- **Do NOT proceed to Step 3 until brainstorming confirms the design is approved.**
42
+ Once you have sufficient clarity, propose **2–3 solution approaches** with trade-offs and a recommendation:
43
+ - Present options conversationally, lead with your recommendation
44
+ - Get developer confirmation on the chosen approach
45
+
46
+ **Do NOT proceed to Step 3 until the developer has confirmed the design direction.**
47
+
48
+ > ⚠️ **DO NOT invoke `superpowers:brainstorming`** — brainstorming's terminal state invokes
49
+ > `writing-plans` (Gate 2), which would bypass Gate 1's approval gate. All Q&A happens here.
41
50
 
42
51
  ---
43
52
 
44
53
  ### Step 3: Extend with Impact Analysis & Effort Estimate
45
54
 
46
- After brainstorming completes, add what brainstorming does not cover:
55
+ After the developer confirms the design direction (Step 2), extend with:
47
56
 
48
57
  **Impact Analysis:**
49
58
  - Which services, modules, and jobs are affected?
@@ -219,8 +228,8 @@ Please review the requirement document.
219
228
 
220
229
  | Concern | Handled by |
221
230
  |---------|-----------|
222
- | Q&A loop (one question at a time) | `superpowers:brainstorming` |
223
- | Propose 2–3 solution approaches | `superpowers:brainstorming` |
231
+ | Q&A loop (one question at a time) | This skill (inline) |
232
+ | Propose 2–3 solution approaches | This skill (inline) |
224
233
  | Root cause tracing (bug tasks) | `superpowers:systematic-debugging` |
225
234
  | Impact analysis | This skill |
226
235
  | Effort estimate | This skill |
@@ -231,10 +240,11 @@ Please review the requirement document.
231
240
 
232
241
  ## Mandatory Rules
233
242
 
234
- - ❌ **DO NOT** skip invoking `superpowers:brainstorming` — it owns the Q&A loop
243
+ - ❌ **DO NOT** invoke `superpowers:brainstorming` — its terminal state calls `writing-plans` and bypasses Gate 1 approval
244
+ - ❌ **DO NOT** invoke `superpowers:writing-plans` — that belongs to Gate 2 (`generate-spec` skill)
235
245
  - ❌ **DO NOT** proceed past Step 2 if design is not confirmed by developer
236
246
  - ❌ **DO NOT** write code at any point in this skill
237
247
  - ❌ **DO NOT** skip impact analysis and effort estimate
238
- - ✅ **MUST** read source code before brainstorming begins
248
+ - ✅ **MUST** read source code before Q&A begins
239
249
  - ✅ **MUST** save `plan/[ticket-id]/requirement.md`
240
250
  - ✅ **MUST** display Gate 1 prompt and wait for `APPROVED`
@@ -128,7 +128,7 @@ Summary: plan/[ticket-id]/summary.md
128
128
 
129
129
  ✅ If OK: type **"APPROVED"**
130
130
  🐛 If there are bugs: type **"BUG: [description]"**
131
- → If a requirement bug: I will return to Gate 2
131
+ → If a requirement bug: I will return to Gate 1
132
132
  → If a coding bug: I will fix and repeat Gate 4
133
133
  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
134
134
  ```
@@ -1,88 +1,88 @@
1
- ## MANDATORY: Strict Gate Workflow
2
-
3
- > **Absolute Rule: Complete Gate N before entering Gate N+1.**
4
- > **Do not skip, shorten, or merge Gates.**
5
-
6
- You have superpowers. When a ticket context exists in `.aiflow/context/current.json`:
7
- - **Consult the "AI Skill Registry"** below to find instructions for each skill (`SKILL.md`).
8
- - **AUTO-START Gate 1 immediately** — do NOT wait for the developer to ask.
9
- - Read instructions and follow the gate sequence below — NO EXCEPTIONS.
10
-
11
- ---
12
-
13
- ### GATE 1 — AI Analyze Requirement (auto-start)
14
-
15
- **INVOKE:** `validate-ticket` skill
16
-
17
- AI actively reads ticket + source code to understand the requirement:
18
- 1. Read `.aiflow/context/current.json` — ticket info
19
- 2. Read source code — architecture, related files, data flow
20
- 3. If anything is unclear — ask ONE question at a time, wait for reply
21
- 4. Output `plan/[ticket-id]/requirement.md` with:
22
- - Requirements summary, source code analysis
23
- - Proposed solution and approach
24
- - Impact analysis, effort estimate, testing plan
25
- 5. Display "GATE 1: Requirement doc ready" → wait for **APPROVED**
26
-
27
- > **Tip:** If auto-start doesn't trigger, the developer can start this gate by typing: **"start"**, **"Gate 1"** or **"Analyze ticket"**.
28
-
29
- DO NOT just check format — **understand the content and propose solutions**.
30
-
31
- > **Telemetry:** Run `aiflow gate 1 start --ticket [ticket-id]` when starting this gate (auto-start or when developer types "start"/"Gate 1").
32
- > Run `aiflow gate 1 approved --ticket [ticket-id]` immediately when APPROVED is received.
33
-
34
- ---
35
-
36
- ### GATE 2 — Implementation Plan (wait for APPROVED)
37
-
38
- **INVOKE:** `generate-spec` skill, then `superpowers:writing-plans`
39
-
40
- - Create a detailed TDD implementation plan based on the approved requirement.
41
- - Display: "GATE 2 PAUSED: type APPROVED to start coding".
42
- - CODE WILL NOT BE GENERATED until "APPROVED" is received.
43
-
44
- > **Telemetry:** Run `aiflow gate 2 start --ticket [ticket-id]` when starting this gate.
45
- > Run `aiflow gate 2 approved --ticket [ticket-id]` immediately when APPROVED is received.
46
-
47
- ---
48
-
49
- ### GATE 3 — Code Generation (TDD only)
50
-
51
- Only runs after Gate 2 has been APPROVED.
52
-
53
- **INVOKE:** `superpowers:test-driven-development`
54
- - Complex feature (3+ files): `superpowers:subagent-driven-development`
55
- - Write tests FIRST — run to confirm FAIL -> implement -> PASS.
56
- - Bug fix EXTRA: `superpowers:systematic-debugging` + `investigate-bug` skill first.
57
-
58
- > **Telemetry:** Run `aiflow gate 3 start --ticket [ticket-id]` when starting this gate.
59
-
60
- ---
61
-
62
- ### GATE 4 — AI Self-Review (wait for APPROVED)
63
-
64
- **INVOKE:** `review-plan` skill
65
-
66
- Mandatory order:
67
- 1. `superpowers:verification-before-completion` — all tests must PASS
68
- 2. `impact-analysis` skill — check for breaking changes
69
- 3. Tick `custom/rules/review-checklist.md`
70
- 4. Create `plan/[ticket-id]/summary.md`
71
-
72
- Then: "GATE 4 PAUSED: type APPROVED or BUG: [description]"
73
- - Coding bug -> fix -> repeat Gate 4.
74
- - Requirement bug -> return to Gate 1.
75
-
76
- > **Telemetry:** Run `aiflow gate 4 start --ticket [ticket-id]` when starting this gate.
77
- > Run `aiflow gate 4 approved --ticket [ticket-id]` immediately when APPROVED is received.
78
-
79
- ---
80
-
81
- ### GATE 5 — Peer Review and Done
82
-
83
- Only runs after Gate 4 has been APPROVED.
84
-
85
- **INVOKE:** `superpowers:requesting-code-review`
86
- Guide on creating a Pull Request with the ticket link.
87
-
88
- > **Telemetry:** Run `aiflow gate 5 start --ticket [ticket-id]` when starting this gate.
1
+ ## MANDATORY: Strict Gate Workflow
2
+
3
+ > **Absolute Rule: Complete Gate N before entering Gate N+1.**
4
+ > **Do not skip, shorten, or merge Gates.**
5
+
6
+ You have superpowers. When a ticket context exists in `.aiflow/context/current.json`:
7
+ - **Consult the "AI Skill Registry"** below to find instructions for each skill (`SKILL.md`).
8
+ - **AUTO-START Gate 1 immediately** — do NOT wait for the developer to ask.
9
+ - Read instructions and follow the gate sequence below — NO EXCEPTIONS.
10
+
11
+ ---
12
+
13
+ ### GATE 1 — AI Analyze Requirement (auto-start)
14
+
15
+ **INVOKE:** `read-study-requirement` skill
16
+
17
+ AI actively reads ticket + source code to understand the requirement:
18
+ 1. Read `.aiflow/context/current.json` — ticket info
19
+ 2. Read source code — architecture, related files, data flow
20
+ 3. If anything is unclear — ask ONE question at a time, wait for reply
21
+ 4. Output `plan/[ticket-id]/requirement.md` with:
22
+ - Requirements summary, source code analysis
23
+ - Proposed solution and approach
24
+ - Impact analysis, effort estimate, testing plan
25
+ 5. Display "GATE 1: Requirement doc ready" → wait for **APPROVED**
26
+
27
+ > **Tip:** If auto-start doesn't trigger, the developer can start this gate by typing: **"start"**, **"Gate 1"** or **"Analyze ticket"**.
28
+
29
+ DO NOT just check format — **understand the content and propose solutions**.
30
+
31
+ > **Telemetry:** Run `aiflow gate 1 start --ticket [ticket-id]` when starting this gate (auto-start or when developer types "start"/"Gate 1").
32
+ > Run `aiflow gate 1 approved --ticket [ticket-id]` immediately when APPROVED is received.
33
+
34
+ ---
35
+
36
+ ### GATE 2 — Implementation Plan (wait for APPROVED)
37
+
38
+ **INVOKE:** `generate-spec` skill, then `superpowers:writing-plans`
39
+
40
+ - Create a detailed TDD implementation plan based on the approved requirement.
41
+ - Display: "GATE 2 PAUSED: type APPROVED to start coding".
42
+ - CODE WILL NOT BE GENERATED until "APPROVED" is received.
43
+
44
+ > **Telemetry:** Run `aiflow gate 2 start --ticket [ticket-id]` when starting this gate.
45
+ > Run `aiflow gate 2 approved --ticket [ticket-id]` immediately when APPROVED is received.
46
+
47
+ ---
48
+
49
+ ### GATE 3 — Code Generation (TDD only)
50
+
51
+ Only runs after Gate 2 has been APPROVED.
52
+
53
+ **INVOKE:** `superpowers:test-driven-development`
54
+ - Complex feature (3+ files): `superpowers:subagent-driven-development`
55
+ - Write tests FIRST — run to confirm FAIL -> implement -> PASS.
56
+ - Bug fix EXTRA: `superpowers:systematic-debugging` + `investigate-bug` skill first.
57
+
58
+ > **Telemetry:** Run `aiflow gate 3 start --ticket [ticket-id]` when starting this gate.
59
+
60
+ ---
61
+
62
+ ### GATE 4 — AI Self-Review (wait for APPROVED)
63
+
64
+ **INVOKE:** `review-plan` skill
65
+
66
+ Mandatory order:
67
+ 1. `superpowers:verification-before-completion` — all tests must PASS
68
+ 2. `impact-analysis` skill — check for breaking changes
69
+ 3. Tick `custom/rules/review-checklist.md`
70
+ 4. Create `plan/[ticket-id]/summary.md`
71
+
72
+ Then: "GATE 4 PAUSED: type APPROVED or BUG: [description]"
73
+ - Coding bug -> fix -> repeat Gate 4.
74
+ - Requirement bug -> return to Gate 1.
75
+
76
+ > **Telemetry:** Run `aiflow gate 4 start --ticket [ticket-id]` when starting this gate.
77
+ > Run `aiflow gate 4 approved --ticket [ticket-id]` immediately when APPROVED is received.
78
+
79
+ ---
80
+
81
+ ### GATE 5 — Peer Review and Done
82
+
83
+ Only runs after Gate 4 has been APPROVED.
84
+
85
+ **INVOKE:** `superpowers:requesting-code-review`
86
+ Guide on creating a Pull Request with the ticket link.
87
+
88
+ > **Telemetry:** Run `aiflow gate 5 start --ticket [ticket-id]` when starting this gate.
@@ -8,6 +8,6 @@ If Gate 1 does not auto-start, wait for the developer to type **"start"**, **"Ga
8
8
 
9
9
  ## Interaction Rules:
10
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.
11
+ - **COLLABORATIVE SKILLS:** When a skill (like `read-study-requirement`) says to "ask one question at a time", you MUST stop and wait for the developer's reply before proceeding.
12
12
  - **NEVER BATCH QUESTIONS:** Only ask one question per message.
13
13
  - **WAIT FOR APPROVAL:** Do not move to the next Gate until you receive "APPROVED".
@@ -7,6 +7,6 @@ You are an expert AI assistant specialized in this project's stack. Follow the G
7
7
  If the Gate Workflow hasn't started, wait for the developer to type **"start"** or **"Gate 1"**.
8
8
 
9
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.
10
+ - **COLLABORATIVE SKILLS:** When a skill (like `read-study-requirement`) says to "ask one question at a time", you MUST stop and wait for the developer's reply before proceeding.
11
11
  - **NEVER BATCH QUESTIONS:** Only ask one question per message.
12
12
  - **WAIT FOR APPROVAL:** Do not move to the next Gate until you receive "APPROVED".
@@ -7,6 +7,6 @@ You are an expert AI assistant specialized in this project's stack. Follow the G
7
7
  If Gate 1 doesn't auto-start, wait for the developer to type **"start"**, **"Gate 1"** or **"Analyze ticket"**.
8
8
 
9
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.
10
+ - **COLLABORATIVE SKILLS:** When a skill (like `read-study-requirement`) says to "ask one question at a time", you MUST stop and wait for the developer's reply before proceeding.
11
11
  - **NEVER BATCH QUESTIONS:** Only ask one question per message.
12
12
  - **WAIT FOR APPROVAL:** Do not move to the next Gate until you receive "APPROVED".
@@ -7,6 +7,6 @@ You are an expert AI assistant specialized in this project's stack. Follow the G
7
7
  If no instructions are automatically followed, wait for the developer to type **"start"** or **"Gate 1"** to start the analysis.
8
8
 
9
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.
10
+ - **COLLABORATIVE SKILLS:** When a skill (like `read-study-requirement`) says to "ask one question at a time", you MUST stop and wait for the developer's reply before proceeding.
11
11
  - **NEVER BATCH QUESTIONS:** Only ask one question per message.
12
12
  - **WAIT FOR APPROVAL:** Do not move to the next Gate until you receive "APPROVED".
@@ -12,6 +12,6 @@ Follow the gated workflow and rules defined below.
12
12
  If Gate 1 does not start automatically, please start it when the developer types **"Gate 1"**.
13
13
 
14
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.
15
+ - **COLLABORATIVE SKILLS:** When a skill (like `read-study-requirement`) says to "ask one question at a time", you MUST stop and wait for the developer's reply before proceeding.
16
16
  - **NEVER BATCH QUESTIONS:** Only ask one question per message.
17
17
  - **WAIT FOR APPROVAL:** Do not move to the next Gate until you receive "APPROVED".