planflow-ai 1.1.6 → 1.1.7
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.
- package/.claude/commands/create-plan.md +57 -0
- package/.claude/commands/discovery-plan.md +75 -0
- package/.claude/commands/execute-plan.md +71 -0
- package/.claude/commands/flow.md +148 -62
- package/.claude/commands/heartbeat.md +6 -1
- package/.claude/commands/review-code.md +68 -0
- package/.claude/commands/review-pr.md +69 -0
- package/.claude/commands/write-tests.md +22 -0
- package/.claude/resources/core/_index.md +12 -2
- package/.claude/resources/core/autopilot-mode.md +11 -8
- package/.claude/resources/core/heartbeat.md +47 -0
- package/.claude/resources/core/learn-recommendations.md +170 -0
- package/.claude/resources/core/project-tasklist.md +160 -2
- package/dist/cli/commands/heartbeat.d.ts +5 -0
- package/dist/cli/commands/heartbeat.d.ts.map +1 -1
- package/dist/cli/commands/heartbeat.js +27 -0
- package/dist/cli/commands/heartbeat.js.map +1 -1
- package/dist/cli/commands/init.d.ts.map +1 -1
- package/dist/cli/commands/init.js +7 -0
- package/dist/cli/commands/init.js.map +1 -1
- package/dist/cli/daemon/heartbeat-daemon.js +51 -0
- package/dist/cli/daemon/heartbeat-daemon.js.map +1 -1
- package/dist/cli/daemon/heartbeat-parser.d.ts +2 -0
- package/dist/cli/daemon/heartbeat-parser.d.ts.map +1 -1
- package/dist/cli/daemon/heartbeat-parser.js +26 -0
- package/dist/cli/daemon/heartbeat-parser.js.map +1 -1
- package/dist/cli/handlers/shared.d.ts.map +1 -1
- package/dist/cli/handlers/shared.js +83 -3
- package/dist/cli/handlers/shared.js.map +1 -1
- package/dist/cli/types.d.ts +3 -1
- package/dist/cli/types.d.ts.map +1 -1
- package/package.json +1 -1
- package/rules/skills/discovery-skill.mdc +2 -0
|
@@ -275,3 +275,60 @@ When executing this command:
|
|
|
275
275
|
| `complexity-scoring.md` | Complexity scoring system |
|
|
276
276
|
| `/discovery-plan` command | Run discovery first |
|
|
277
277
|
| `/execute-plan` command | Execute the created plan |
|
|
278
|
+
|
|
279
|
+
---
|
|
280
|
+
|
|
281
|
+
## Tasklist Updates
|
|
282
|
+
|
|
283
|
+
Update `flow/tasklist.md` at these points. See `.claude/resources/core/project-tasklist.md` for full rules.
|
|
284
|
+
|
|
285
|
+
1. **On start**: Add "Plan: {feature}" to **In Progress** (or move it from To Do if it already exists)
|
|
286
|
+
2. **On complete**: Move "Plan: {feature}" to **Done** with today's date
|
|
287
|
+
3. **Next step**: Add "Execute plan for {feature}" to **To Do**
|
|
288
|
+
|
|
289
|
+
---
|
|
290
|
+
|
|
291
|
+
## Learn Recommendations
|
|
292
|
+
|
|
293
|
+
After plan creation completes, check for learning opportunities. See `.claude/resources/core/learn-recommendations.md` for the full system.
|
|
294
|
+
|
|
295
|
+
**Plan-specific checks**:
|
|
296
|
+
- Does the plan introduce **new technologies** not in the current tech stack?
|
|
297
|
+
- Are there phases that require **unfamiliar APIs or services**?
|
|
298
|
+
- Did the discovery document flag any "Suggested Learning" topics?
|
|
299
|
+
|
|
300
|
+
Present recommendations after the plan summary.
|
|
301
|
+
|
|
302
|
+
---
|
|
303
|
+
|
|
304
|
+
## Brain Capture
|
|
305
|
+
|
|
306
|
+
After plan creation completes successfully, append a brain-capture block. See `.claude/resources/core/brain-capture.md` for processing rules.
|
|
307
|
+
|
|
308
|
+
**Capture the following**:
|
|
309
|
+
|
|
310
|
+
```
|
|
311
|
+
<!-- brain-capture
|
|
312
|
+
skill: create-plan
|
|
313
|
+
feature: [feature name]
|
|
314
|
+
status: completed
|
|
315
|
+
data:
|
|
316
|
+
phase_count: [number of phases]
|
|
317
|
+
total_complexity: [sum of complexity scores]
|
|
318
|
+
highest_phase: [phase name with highest score]
|
|
319
|
+
discovery_link: [[discovery-feature-name]]
|
|
320
|
+
plan_doc: [path to plan document]
|
|
321
|
+
-->
|
|
322
|
+
```
|
|
323
|
+
|
|
324
|
+
Update `flow/brain/features/[feature-name].md` with plan details and link to discovery entry.
|
|
325
|
+
|
|
326
|
+
---
|
|
327
|
+
|
|
328
|
+
## Resource Capture
|
|
329
|
+
|
|
330
|
+
During this skill's execution, watch for valuable reference materials worth preserving. See `.claude/resources/core/resource-capture.md` for capture rules, file format, and naming conventions.
|
|
331
|
+
|
|
332
|
+
At natural break points, if you encounter information that could be useful for future development (API specs, architecture notes, config references, domain knowledge, etc.), ask the user: "I found something that could be useful for future reference: _{brief description}_. Should I save it to `flow/resources/`?"
|
|
333
|
+
|
|
334
|
+
Only save if the user approves. Do not re-ask if declined.
|
|
@@ -305,3 +305,78 @@ Read: resources/tools/interactive-questions-tool.md (lines from TLS-IQ-3)
|
|
|
305
305
|
| `discovery-templates.md` | Document templates |
|
|
306
306
|
| `interactive-questions-tool.md` | Interactive Questions UI workflow |
|
|
307
307
|
| `/create-plan` command | Creates plan from discovery document |
|
|
308
|
+
|
|
309
|
+
---
|
|
310
|
+
|
|
311
|
+
## Tasklist Updates
|
|
312
|
+
|
|
313
|
+
Update `flow/tasklist.md` at these points. See `.claude/resources/core/project-tasklist.md` for full rules.
|
|
314
|
+
|
|
315
|
+
1. **On start**: Add "Discovery: {feature}" to **In Progress** (or move it from To Do if it already exists)
|
|
316
|
+
2. **On complete**: Move "Discovery: {feature}" to **Done** with today's date
|
|
317
|
+
3. **Next step**: Add "Create plan for {feature}" to **To Do**
|
|
318
|
+
|
|
319
|
+
---
|
|
320
|
+
|
|
321
|
+
## Learn Recommendations
|
|
322
|
+
|
|
323
|
+
After discovery completes, check for learning opportunities. See `.claude/resources/core/learn-recommendations.md` for the full system.
|
|
324
|
+
|
|
325
|
+
**Discovery-specific checks**:
|
|
326
|
+
- Does the proposed approach introduce **new technologies** the project hasn't used before?
|
|
327
|
+
- Are there **unfamiliar APIs or services** in the requirements?
|
|
328
|
+
- Did the user mention switching from one pattern/tool to another?
|
|
329
|
+
|
|
330
|
+
**If new tech is detected**, add a "Suggested Learning" section to the discovery document:
|
|
331
|
+
|
|
332
|
+
```markdown
|
|
333
|
+
## Suggested Learning
|
|
334
|
+
|
|
335
|
+
The following topics are new to this project and may benefit from structured learning before implementation:
|
|
336
|
+
|
|
337
|
+
| Topic | Why | Suggested Command |
|
|
338
|
+
|-------|-----|-------------------|
|
|
339
|
+
| Langfuse | New observability tool not previously used | `/learn langfuse` |
|
|
340
|
+
| WebSocket | Real-time communication pattern new to project | `/learn websockets` |
|
|
341
|
+
|
|
342
|
+
These are optional but recommended to reduce implementation friction.
|
|
343
|
+
```
|
|
344
|
+
|
|
345
|
+
**Detection**: Compare proposed technologies against `flow/references/tech-foundation.md` and existing `package.json` / `pyproject.toml` dependencies. If a technology is not in the current stack, flag it.
|
|
346
|
+
|
|
347
|
+
**Presentation**: Include the section in the discovery document itself (not as a chat message). This ensures the learning suggestion persists and is visible when the user reviews the document.
|
|
348
|
+
|
|
349
|
+
---
|
|
350
|
+
|
|
351
|
+
## Brain Capture
|
|
352
|
+
|
|
353
|
+
After discovery completes successfully, append a brain-capture block. See `.claude/resources/core/brain-capture.md` for processing rules.
|
|
354
|
+
|
|
355
|
+
**Capture the following**:
|
|
356
|
+
|
|
357
|
+
```
|
|
358
|
+
<!-- brain-capture
|
|
359
|
+
skill: discovery
|
|
360
|
+
feature: [feature name from discovery]
|
|
361
|
+
status: completed
|
|
362
|
+
data:
|
|
363
|
+
user_prompt: [original user request]
|
|
364
|
+
questions_asked: [count]
|
|
365
|
+
questions_answered: [count]
|
|
366
|
+
requirements_fr: [count of functional requirements]
|
|
367
|
+
requirements_nfr: [count of non-functional requirements]
|
|
368
|
+
discovery_doc: [path to discovery document]
|
|
369
|
+
-->
|
|
370
|
+
```
|
|
371
|
+
|
|
372
|
+
Write/update `flow/brain/features/[feature-name].md` with discovery context and update `flow/brain/index.md`.
|
|
373
|
+
|
|
374
|
+
---
|
|
375
|
+
|
|
376
|
+
## Resource Capture
|
|
377
|
+
|
|
378
|
+
During this skill's execution, watch for valuable reference materials worth preserving. See `.claude/resources/core/resource-capture.md` for capture rules, file format, and naming conventions.
|
|
379
|
+
|
|
380
|
+
At natural break points, if you encounter information that could be useful for future development (API specs, architecture notes, config references, domain knowledge, etc.), ask the user: "I found something that could be useful for future reference: _{brief description}_. Should I save it to `flow/resources/`?"
|
|
381
|
+
|
|
382
|
+
Only save if the user approves. Do not re-ask if declined.
|
|
@@ -140,6 +140,53 @@ Please run this command and let me know when it's complete.
|
|
|
140
140
|
|
|
141
141
|
---
|
|
142
142
|
|
|
143
|
+
## Git Control
|
|
144
|
+
|
|
145
|
+
On execution start, check if `flow/.flowconfig` exists — if yes, read git control settings (`commit`, `push`, `branch`) from it. Fallback: check `flow/.gitcontrol` for backward compatibility. If neither exists, no git operations are performed.
|
|
146
|
+
|
|
147
|
+
### Git Control Settings
|
|
148
|
+
|
|
149
|
+
```yaml
|
|
150
|
+
# flow/.gitcontrol
|
|
151
|
+
commit: true # Auto-commit after each completed phase
|
|
152
|
+
push: true # Auto-push after all phases complete (requires commit: true)
|
|
153
|
+
branch: develop # Target branch (optional, defaults to current branch)
|
|
154
|
+
```
|
|
155
|
+
|
|
156
|
+
### Git Behavior During Execution
|
|
157
|
+
|
|
158
|
+
| Setting | Behavior |
|
|
159
|
+
|---------|----------|
|
|
160
|
+
| `commit: true` | After each phase completes successfully, run `git add -A && git commit -m "Phase N: <phase name> — <feature>"` |
|
|
161
|
+
| `commit: false` or no `.gitcontrol` | No automatic git operations (default behavior) |
|
|
162
|
+
| `push: true` | After ALL phases complete AND build/test pass, run `git push origin <branch>` |
|
|
163
|
+
| `push: false` or not set | No automatic push |
|
|
164
|
+
| `branch: <name>` | Use this branch for push (default: current branch) |
|
|
165
|
+
|
|
166
|
+
### Git Safety Rules
|
|
167
|
+
|
|
168
|
+
| Rule | Description |
|
|
169
|
+
|------|-------------|
|
|
170
|
+
| **Commit only on success** | Only commit after a phase completes successfully. Never commit broken code. |
|
|
171
|
+
| **Push only after build+test** | Push only after `npm run build && npm run test` pass at the very end. |
|
|
172
|
+
| **No force push** | NEVER use `--force`. If push fails, stop and ask the user. |
|
|
173
|
+
| **Commit message format** | `Phase N: <phase name> — <feature>` (e.g., "Phase 2: API endpoints — user-auth") |
|
|
174
|
+
| **Final commit** | After build+test pass, make one final commit: `Complete: <feature> — all phases done, build passing` |
|
|
175
|
+
| **Include flow artifacts** | Commits should include updated plan files with progress markers |
|
|
176
|
+
|
|
177
|
+
### Example Flow with `commit=true push=true`
|
|
178
|
+
|
|
179
|
+
```
|
|
180
|
+
Phase 1: Setup types → completes → git commit "Phase 1: Setup types — user-auth"
|
|
181
|
+
Phase 2: API endpoints → completes → git commit "Phase 2: API endpoints — user-auth"
|
|
182
|
+
Phase 3: Frontend UI → completes → git commit "Phase 3: Frontend UI — user-auth"
|
|
183
|
+
Phase 4: Tests → completes → git commit "Phase 4: Tests — user-auth"
|
|
184
|
+
Build + Test → pass → git commit "Complete: user-auth — all phases done, build passing"
|
|
185
|
+
→ git push origin development
|
|
186
|
+
```
|
|
187
|
+
|
|
188
|
+
---
|
|
189
|
+
|
|
143
190
|
## Instructions
|
|
144
191
|
|
|
145
192
|
### Step 1: Validate Inputs
|
|
@@ -351,6 +398,30 @@ Update `flow/brain/features/[feature-name].md` with execution results. If errors
|
|
|
351
398
|
|
|
352
399
|
---
|
|
353
400
|
|
|
401
|
+
## Tasklist Updates
|
|
402
|
+
|
|
403
|
+
Update `flow/tasklist.md` at these points. See `.claude/resources/core/project-tasklist.md` for full rules.
|
|
404
|
+
|
|
405
|
+
1. **On start**: Add "Execute: {feature}" to **In Progress** (or move it from To Do if it already exists)
|
|
406
|
+
2. **On complete**: Move "Execute: {feature}" to **Done** with today's date
|
|
407
|
+
3. **Next step**: Add "Review code for {feature}" to **To Do**
|
|
408
|
+
|
|
409
|
+
---
|
|
410
|
+
|
|
411
|
+
## Learn Recommendations
|
|
412
|
+
|
|
413
|
+
After execution completes (or at the end of each phase), check for learning opportunities. See `.claude/resources/core/learn-recommendations.md` for the full system.
|
|
414
|
+
|
|
415
|
+
**Quick checks at phase end**:
|
|
416
|
+
- Were new dependencies added? → Recommend `/learn <dependency>`
|
|
417
|
+
- Were non-trivial errors resolved (3+ attempts)? → Recommend `/learn` (pattern extraction)
|
|
418
|
+
- Did the user correct the approach? → Recommend `/learn <corrected-topic>`
|
|
419
|
+
- Was a new technology/pattern introduced? → Recommend `/learn <topic>`
|
|
420
|
+
|
|
421
|
+
**Present batched recommendations** at the end of execution, before the final summary. Do NOT interrupt phase execution.
|
|
422
|
+
|
|
423
|
+
---
|
|
424
|
+
|
|
354
425
|
## Resource Capture
|
|
355
426
|
|
|
356
427
|
During this skill's execution, watch for valuable reference materials worth preserving. See `.claude/resources/core/resource-capture.md` for capture rules, file format, and naming conventions.
|
package/.claude/commands/flow.md
CHANGED
|
@@ -1,16 +1,17 @@
|
|
|
1
1
|
---
|
|
2
|
-
description:
|
|
2
|
+
description: Configure plan-flow settings — autopilot mode, git control, and other runtime options. Use key=value syntax.
|
|
3
3
|
---
|
|
4
4
|
|
|
5
|
-
# Flow:
|
|
5
|
+
# Flow: Plan-Flow Configuration
|
|
6
6
|
|
|
7
7
|
## Command Description
|
|
8
8
|
|
|
9
|
-
This command
|
|
9
|
+
This command is the **central configuration hub** for plan-flow runtime settings. All settings use `key=value` syntax and are persisted in `flow/.flowconfig` (YAML format).
|
|
10
10
|
|
|
11
|
-
|
|
12
|
-
|
|
13
|
-
|
|
11
|
+
Settings include:
|
|
12
|
+
- **Autopilot mode** — auto-chain commands for feature requests
|
|
13
|
+
- **Git control** — auto-commit per phase, auto-push after completion
|
|
14
|
+
- **Branch targeting** — specify which branch to commit/push to
|
|
14
15
|
|
|
15
16
|
---
|
|
16
17
|
|
|
@@ -19,109 +20,194 @@ The LLM pauses only at mandatory checkpoints (discovery Q&A, plan approval).
|
|
|
19
20
|
**If the user invokes this command with `-help`, display only this section and stop:**
|
|
20
21
|
|
|
21
22
|
```
|
|
22
|
-
/flow -
|
|
23
|
+
/flow - Plan-Flow Configuration
|
|
23
24
|
|
|
24
25
|
DESCRIPTION:
|
|
25
|
-
|
|
26
|
-
|
|
26
|
+
Central configuration command for plan-flow runtime settings.
|
|
27
|
+
All settings use key=value syntax and persist across sessions.
|
|
27
28
|
|
|
28
29
|
USAGE:
|
|
29
|
-
/flow <
|
|
30
|
-
/flow
|
|
31
|
-
/flow -
|
|
32
|
-
/flow -
|
|
33
|
-
/flow
|
|
34
|
-
|
|
30
|
+
/flow <key>=<value> [key=value ...] Set one or more settings
|
|
31
|
+
/flow <prompt> Enable autopilot and start flow with prompt
|
|
32
|
+
/flow -status Show all current settings
|
|
33
|
+
/flow -reset Reset all settings to defaults
|
|
34
|
+
/flow -help Show this help
|
|
35
|
+
|
|
36
|
+
SETTINGS:
|
|
37
|
+
autopilot=true|false Enable/disable autopilot mode (default: false)
|
|
38
|
+
commit=true|false Auto-commit after each completed phase (default: false)
|
|
39
|
+
push=true|false Auto-push after all phases + build/test pass (default: false)
|
|
40
|
+
branch=<name> Target branch for git operations (default: current branch)
|
|
35
41
|
|
|
36
42
|
EXAMPLES:
|
|
37
|
-
/flow
|
|
38
|
-
/flow
|
|
39
|
-
/flow
|
|
40
|
-
|
|
41
|
-
|
|
43
|
+
/flow autopilot=true # Enable autopilot
|
|
44
|
+
/flow commit=true push=true # Enable git control (works without autopilot)
|
|
45
|
+
/flow autopilot=true commit=true # Enable both
|
|
46
|
+
/flow branch=development # Set target branch
|
|
47
|
+
/flow commit=false push=false # Disable git control
|
|
48
|
+
/flow -status # Show current config
|
|
49
|
+
/flow -reset # Reset everything
|
|
50
|
+
|
|
51
|
+
# Shorthand: text without key=value enables autopilot and starts flow
|
|
52
|
+
/flow add dark mode support # autopilot=true + start discovery
|
|
53
|
+
/flow commit=true add user auth # autopilot=true + git + start discovery
|
|
54
|
+
|
|
55
|
+
BEHAVIOR WHEN AUTOPILOT IS ON:
|
|
42
56
|
- Feature requests → full flow (discovery → plan → execute → review → archive)
|
|
43
57
|
- Trivial tasks (complexity 0-2) → executed directly, no flow
|
|
44
58
|
- Questions/exploration → answered normally
|
|
45
59
|
- Slash commands → run as normal
|
|
46
60
|
|
|
61
|
+
GIT CONTROL (when commit=true):
|
|
62
|
+
- After each phase completes: auto-commit
|
|
63
|
+
"Phase N: <phase name> — <feature>"
|
|
64
|
+
- After all phases + build/test pass: auto-push if push=true
|
|
65
|
+
- If build/test fails: commit is made but push is skipped
|
|
66
|
+
- push=true without commit=true: auto-enables commit=true
|
|
67
|
+
|
|
47
68
|
MANDATORY CHECKPOINTS (even in autopilot):
|
|
48
69
|
- Discovery phase: pauses for user Q&A
|
|
49
70
|
- Plan review: pauses for user approval before execution
|
|
50
71
|
|
|
51
|
-
|
|
52
|
-
|
|
53
|
-
|
|
54
|
-
|
|
55
|
-
STATE:
|
|
56
|
-
Persisted in flow/.autopilot (survives session restarts)
|
|
72
|
+
CONFIG FILE:
|
|
73
|
+
All settings stored in flow/.flowconfig (YAML)
|
|
74
|
+
Backward compatible: also reads flow/.autopilot and flow/.gitcontrol
|
|
57
75
|
```
|
|
58
76
|
|
|
59
77
|
---
|
|
60
78
|
|
|
61
79
|
## Instructions
|
|
62
80
|
|
|
63
|
-
###
|
|
81
|
+
### Step 1: Parse Input
|
|
64
82
|
|
|
65
|
-
|
|
66
|
-
2. Confirm to the user:
|
|
83
|
+
Parse the user input to determine what action to take:
|
|
67
84
|
|
|
68
|
-
|
|
69
|
-
|
|
85
|
+
1. **Scan for `key=value` pairs**: Extract all `key=value` tokens from the input
|
|
86
|
+
2. **Check for flags**: `-status`, `-reset`, `-help`
|
|
87
|
+
3. **Remaining text**: Anything left after extracting keys and flags is a prompt
|
|
70
88
|
|
|
71
|
-
|
|
72
|
-
1. Check contracts → 2. Discovery (pause for Q&A) → 3. Create plan (pause for approval) → 4. Execute plan → 5. Review code → 6. Archive
|
|
89
|
+
**Valid keys**:
|
|
73
90
|
|
|
74
|
-
|
|
91
|
+
| Key | Values | Default | Description |
|
|
92
|
+
|-----|--------|---------|-------------|
|
|
93
|
+
| `autopilot` | `true`, `false` | `false` | Enable/disable autopilot mode |
|
|
94
|
+
| `commit` | `true`, `false` | `false` | Auto-commit after each phase |
|
|
95
|
+
| `push` | `true`, `false` | `false` | Auto-push after completion |
|
|
96
|
+
| `branch` | any string | current branch | Target branch for git ops |
|
|
75
97
|
|
|
76
|
-
|
|
77
|
-
```
|
|
98
|
+
---
|
|
78
99
|
|
|
79
|
-
###
|
|
100
|
+
### Step 2: Load Existing Config
|
|
80
101
|
|
|
81
|
-
1.
|
|
82
|
-
2.
|
|
102
|
+
1. If `flow/.flowconfig` exists, read it
|
|
103
|
+
2. **Backward compatibility**: If `.flowconfig` doesn't exist but `flow/.autopilot` does, read it and migrate:
|
|
104
|
+
- `flow/.autopilot` exists → `autopilot: true`
|
|
105
|
+
- `flow/.gitcontrol` exists → read its `commit`, `push`, `branch` values
|
|
106
|
+
3. Merge parsed key=value pairs into existing config (new values override old)
|
|
83
107
|
|
|
84
|
-
|
|
85
|
-
Autopilot flow mode **disabled**.
|
|
108
|
+
---
|
|
86
109
|
|
|
87
|
-
|
|
88
|
-
`/discovery-plan` → `/create-plan` → `/execute-plan` → `/review-code`
|
|
89
|
-
```
|
|
110
|
+
### Step 3: Validate Settings
|
|
90
111
|
|
|
91
|
-
|
|
112
|
+
1. If `push=true` but `commit` is not `true`, auto-enable `commit=true` and warn:
|
|
113
|
+
> `push=true` requires `commit=true`. Enabling auto-commit as well.
|
|
114
|
+
2. If `autopilot=false` and `commit=false` and no other settings, consider removing `.flowconfig`
|
|
115
|
+
|
|
116
|
+
---
|
|
92
117
|
|
|
93
|
-
|
|
118
|
+
### Step 4: Save Config
|
|
94
119
|
|
|
95
|
-
|
|
96
|
-
2. Confirm autopilot is enabled (brief, one line)
|
|
97
|
-
3. **Immediately start the flow** by invoking `/discovery-plan` with the user's prompt as the feature description
|
|
120
|
+
Write the merged config to `flow/.flowconfig`:
|
|
98
121
|
|
|
99
|
-
|
|
100
|
-
|
|
101
|
-
|
|
122
|
+
```yaml
|
|
123
|
+
# flow/.flowconfig — plan-flow runtime settings
|
|
124
|
+
autopilot: true
|
|
125
|
+
commit: true
|
|
126
|
+
push: true
|
|
127
|
+
branch: development
|
|
128
|
+
```
|
|
102
129
|
|
|
103
|
-
|
|
130
|
+
**Also maintain backward-compatible marker files**:
|
|
131
|
+
- If `autopilot: true` → create `flow/.autopilot` (for backward compat with existing rules)
|
|
132
|
+
- If `autopilot: false` → delete `flow/.autopilot`
|
|
133
|
+
- Write git settings to `flow/.gitcontrol` as well (for backward compat)
|
|
104
134
|
|
|
105
135
|
---
|
|
106
136
|
|
|
107
|
-
###
|
|
137
|
+
### Step 5: Handle Action
|
|
138
|
+
|
|
139
|
+
#### If `-status` or no arguments and no key=value pairs
|
|
140
|
+
|
|
141
|
+
Show current config:
|
|
142
|
+
|
|
143
|
+
```markdown
|
|
144
|
+
**Plan-Flow Configuration**
|
|
145
|
+
|
|
146
|
+
| Setting | Value |
|
|
147
|
+
|---------|-------|
|
|
148
|
+
| Autopilot | enabled/disabled |
|
|
149
|
+
| Auto-commit | true/false |
|
|
150
|
+
| Auto-push | true/false |
|
|
151
|
+
| Branch | <name> or (current) |
|
|
152
|
+
```
|
|
153
|
+
|
|
154
|
+
#### If `-reset`
|
|
155
|
+
|
|
156
|
+
1. Delete `flow/.flowconfig`, `flow/.autopilot`, `flow/.gitcontrol`
|
|
157
|
+
2. Confirm: "All plan-flow settings reset to defaults."
|
|
108
158
|
|
|
109
|
-
|
|
110
|
-
|
|
159
|
+
#### If only key=value pairs (no prompt text)
|
|
160
|
+
|
|
161
|
+
1. Save settings (Steps 2-4)
|
|
162
|
+
2. Confirm what changed:
|
|
111
163
|
|
|
112
164
|
```markdown
|
|
113
|
-
|
|
165
|
+
**Settings updated:**
|
|
166
|
+
- commit: false → **true**
|
|
167
|
+
- push: false → **true**
|
|
114
168
|
|
|
115
|
-
|
|
116
|
-
[If disabled]: Use individual slash commands to run each step manually.
|
|
169
|
+
These apply to all `/execute-plan` runs.
|
|
117
170
|
```
|
|
118
171
|
|
|
172
|
+
#### If prompt text remains (with or without key=value pairs)
|
|
173
|
+
|
|
174
|
+
This is a **feature request with autopilot**:
|
|
175
|
+
|
|
176
|
+
1. Set `autopilot: true` (if not already)
|
|
177
|
+
2. Save all settings (Steps 2-4)
|
|
178
|
+
3. Confirm settings briefly (one line)
|
|
179
|
+
4. **Immediately start the flow** by invoking `/discovery-plan` with the prompt
|
|
180
|
+
|
|
181
|
+
**Examples**:
|
|
182
|
+
- `/flow add dark mode` → `autopilot=true`, start discovery
|
|
183
|
+
- `/flow commit=true push=true add user auth` → `autopilot=true, commit=true, push=true`, start discovery
|
|
184
|
+
|
|
119
185
|
---
|
|
120
186
|
|
|
121
187
|
## Critical Rules
|
|
122
188
|
|
|
123
189
|
| Rule | Description |
|
|
124
190
|
| --- | --- |
|
|
125
|
-
| **
|
|
126
|
-
| **
|
|
127
|
-
| **
|
|
191
|
+
| **Single config file** | All settings in `flow/.flowconfig`. Maintain `.autopilot` and `.gitcontrol` for backward compat. |
|
|
192
|
+
| **Key=value syntax** | All settings use `key=value`. No `-enable`/`-disable` flags needed (use `autopilot=true/false`). |
|
|
193
|
+
| **Settings are independent** | `commit=true` works without `autopilot=true`. Git control applies to any `/execute-plan` run. |
|
|
194
|
+
| **Complete and stop** | After updating settings, STOP and wait for user input (unless a prompt was provided). |
|
|
195
|
+
| **Backward compatible** | Still reads `flow/.autopilot` and `flow/.gitcontrol` if `.flowconfig` doesn't exist yet. |
|
|
196
|
+
|
|
197
|
+
---
|
|
198
|
+
|
|
199
|
+
## Config File Format
|
|
200
|
+
|
|
201
|
+
```yaml
|
|
202
|
+
# flow/.flowconfig — plan-flow runtime settings
|
|
203
|
+
# Managed by /flow command. Do not edit manually.
|
|
204
|
+
|
|
205
|
+
autopilot: false
|
|
206
|
+
commit: false
|
|
207
|
+
push: false
|
|
208
|
+
branch: ""
|
|
209
|
+
```
|
|
210
|
+
|
|
211
|
+
**Location**: `flow/.flowconfig`
|
|
212
|
+
**Format**: YAML
|
|
213
|
+
**Persists**: Across sessions (file-based)
|
|
@@ -46,8 +46,13 @@ EXAMPLE TASKS:
|
|
|
46
46
|
every 6 hours - check for new issues and update tasklist
|
|
47
47
|
weekly on Monday at 9:00 AM - generate weekly summary report
|
|
48
48
|
|
|
49
|
+
TASKLIST INTEGRATION:
|
|
50
|
+
Tasks added to the tasklist with a schedule (e.g., "in 1 hour", "at 3 PM")
|
|
51
|
+
automatically create a linked heartbeat task. Both files cross-reference
|
|
52
|
+
each other with [[]] links for Obsidian navigation.
|
|
53
|
+
|
|
49
54
|
RELATED COMMANDS:
|
|
50
|
-
/flow
|
|
55
|
+
/flow Configure plan-flow settings
|
|
51
56
|
/brain Manual brain entry
|
|
52
57
|
```
|
|
53
58
|
|
|
@@ -170,3 +170,71 @@ When executing this command:
|
|
|
170
170
|
| `resources/core/_index.md` | Index of core rules with reference codes |
|
|
171
171
|
| `review-code-skill.md` | Skill that executes the review |
|
|
172
172
|
| `review-pr-patterns.md` | Review patterns and guidelines |
|
|
173
|
+
|
|
174
|
+
---
|
|
175
|
+
|
|
176
|
+
## Tasklist Updates
|
|
177
|
+
|
|
178
|
+
Update `flow/tasklist.md` at these points. See `.claude/resources/core/project-tasklist.md` for full rules.
|
|
179
|
+
|
|
180
|
+
1. **On start**: Add "Review: {scope}" to **In Progress** (or move it from To Do if it already exists)
|
|
181
|
+
2. **On complete**: Move "Review: {scope}" to **Done** with today's date
|
|
182
|
+
|
|
183
|
+
---
|
|
184
|
+
|
|
185
|
+
## Learn Recommendations
|
|
186
|
+
|
|
187
|
+
After the review completes, check for learning opportunities. See `.claude/resources/core/learn-recommendations.md` for the full system.
|
|
188
|
+
|
|
189
|
+
**Review-specific checks** (scan the diff for):
|
|
190
|
+
- **New imports** from packages not previously used in the project
|
|
191
|
+
- **Pattern changes** (different state management, different API patterns, new testing approaches)
|
|
192
|
+
- **New configuration files** for tools not previously in the project (Docker, CI/CD, etc.)
|
|
193
|
+
|
|
194
|
+
Present recommendations at the end of the review document, after the summary:
|
|
195
|
+
|
|
196
|
+
```markdown
|
|
197
|
+
## Learn Opportunities
|
|
198
|
+
|
|
199
|
+
The following new technologies/patterns were detected in this changeset:
|
|
200
|
+
|
|
201
|
+
- `/learn zustand` — New state management library introduced (replaces Redux)
|
|
202
|
+
- `/learn` — Non-trivial error patterns worth capturing as reusable knowledge
|
|
203
|
+
|
|
204
|
+
Run any of these to build structured understanding before or after merging.
|
|
205
|
+
```
|
|
206
|
+
|
|
207
|
+
---
|
|
208
|
+
|
|
209
|
+
## Brain Capture
|
|
210
|
+
|
|
211
|
+
After code review completes, append a brain-capture block. See `.claude/resources/core/brain-capture.md` for processing rules.
|
|
212
|
+
|
|
213
|
+
**Capture the following**:
|
|
214
|
+
|
|
215
|
+
```
|
|
216
|
+
<!-- brain-capture
|
|
217
|
+
skill: review-code
|
|
218
|
+
feature: [feature or scope reviewed]
|
|
219
|
+
status: completed
|
|
220
|
+
data:
|
|
221
|
+
files_reviewed: [count]
|
|
222
|
+
issues_found: [total count]
|
|
223
|
+
severity_critical: [count]
|
|
224
|
+
severity_warning: [count]
|
|
225
|
+
severity_info: [count]
|
|
226
|
+
pattern_conflicts: [list of pattern conflicts found, if any]
|
|
227
|
+
-->
|
|
228
|
+
```
|
|
229
|
+
|
|
230
|
+
Update `flow/brain/features/[feature-name].md` if reviewing a known feature. Log to `flow/brain/sessions/YYYY-MM-DD.md` otherwise.
|
|
231
|
+
|
|
232
|
+
---
|
|
233
|
+
|
|
234
|
+
## Resource Capture
|
|
235
|
+
|
|
236
|
+
During this skill's execution, watch for valuable reference materials worth preserving. See `.claude/resources/core/resource-capture.md` for capture rules, file format, and naming conventions.
|
|
237
|
+
|
|
238
|
+
At natural break points, if you encounter information that could be useful for future development (API specs, architecture notes, config references, domain knowledge, etc.), ask the user: "I found something that could be useful for future reference: _{brief description}_. Should I save it to `flow/resources/`?"
|
|
239
|
+
|
|
240
|
+
Only save if the user approves. Do not re-ask if declined.
|
|
@@ -169,3 +169,72 @@ When executing this command:
|
|
|
169
169
|
| `review-pr-skill.md` | Skill that executes the review |
|
|
170
170
|
| `auth-pr-tool.md` | Authentication tool for PR platforms |
|
|
171
171
|
| `review-pr-patterns.md` | Review patterns and guidelines |
|
|
172
|
+
|
|
173
|
+
---
|
|
174
|
+
|
|
175
|
+
## Tasklist Updates
|
|
176
|
+
|
|
177
|
+
Update `flow/tasklist.md` at these points. See `.claude/resources/core/project-tasklist.md` for full rules.
|
|
178
|
+
|
|
179
|
+
1. **On start**: Add "Review PR: #{number}" to **In Progress** (or move it from To Do if it already exists)
|
|
180
|
+
2. **On complete**: Move "Review PR: #{number}" to **Done** with today's date
|
|
181
|
+
|
|
182
|
+
---
|
|
183
|
+
|
|
184
|
+
## Learn Recommendations
|
|
185
|
+
|
|
186
|
+
After the PR review completes, check for learning opportunities. See `.claude/resources/core/learn-recommendations.md` for the full system.
|
|
187
|
+
|
|
188
|
+
**PR review-specific checks** (scan the diff for):
|
|
189
|
+
- **New imports** from packages not previously used in the project
|
|
190
|
+
- **Pattern changes** (different state management, different API patterns, new testing approaches)
|
|
191
|
+
- **New configuration files** for tools not previously in the project (Docker, CI/CD, etc.)
|
|
192
|
+
- **Framework migrations** (major version upgrades, framework switches)
|
|
193
|
+
|
|
194
|
+
Present recommendations at the end of the review document, after the summary:
|
|
195
|
+
|
|
196
|
+
```markdown
|
|
197
|
+
## Learn Opportunities
|
|
198
|
+
|
|
199
|
+
The following new technologies/patterns were detected in this PR:
|
|
200
|
+
|
|
201
|
+
- `/learn zustand` — New state management library introduced (replaces Redux)
|
|
202
|
+
- `/learn docker` — Docker configuration added to the project for the first time
|
|
203
|
+
|
|
204
|
+
Run any of these to build structured understanding before or after merging.
|
|
205
|
+
```
|
|
206
|
+
|
|
207
|
+
---
|
|
208
|
+
|
|
209
|
+
## Brain Capture
|
|
210
|
+
|
|
211
|
+
After PR review completes, append a brain-capture block. See `.claude/resources/core/brain-capture.md` for processing rules.
|
|
212
|
+
|
|
213
|
+
**Capture the following**:
|
|
214
|
+
|
|
215
|
+
```
|
|
216
|
+
<!-- brain-capture
|
|
217
|
+
skill: review-pr
|
|
218
|
+
feature: [PR title or scope]
|
|
219
|
+
status: completed
|
|
220
|
+
data:
|
|
221
|
+
pr_number: [PR number]
|
|
222
|
+
pr_platform: [github/azure-devops]
|
|
223
|
+
issues_found: [total count]
|
|
224
|
+
review_outcome: [approved/changes-requested/commented]
|
|
225
|
+
severity_critical: [count]
|
|
226
|
+
severity_warning: [count]
|
|
227
|
+
-->
|
|
228
|
+
```
|
|
229
|
+
|
|
230
|
+
Log to `flow/brain/sessions/YYYY-MM-DD.md` with PR review summary.
|
|
231
|
+
|
|
232
|
+
---
|
|
233
|
+
|
|
234
|
+
## Resource Capture
|
|
235
|
+
|
|
236
|
+
During this skill's execution, watch for valuable reference materials worth preserving. See `.claude/resources/core/resource-capture.md` for capture rules, file format, and naming conventions.
|
|
237
|
+
|
|
238
|
+
At natural break points, if you encounter information that could be useful for future development (API specs, architecture notes, config references, domain knowledge, etc.), ask the user: "I found something that could be useful for future reference: _{brief description}_. Should I save it to `flow/resources/`?"
|
|
239
|
+
|
|
240
|
+
Only save if the user approves. Do not re-ask if declined.
|