wb-flow 1.0.0-r01

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 (48) hide show
  1. package/CHANGELOG.md +12 -0
  2. package/LICENSE +21 -0
  3. package/README.md +128 -0
  4. package/assets/demo.gif +0 -0
  5. package/bin/install.js +175 -0
  6. package/bin/link.js +71 -0
  7. package/bin/verify-wrappers.js +49 -0
  8. package/package.json +56 -0
  9. package/templates/commands/_shared/output_conventions.md +433 -0
  10. package/templates/commands/_shared/wb_universal_agent_instructions.md +72 -0
  11. package/templates/commands/model_recommendation_updates.md +74 -0
  12. package/templates/commands/model_recommendations.md +112 -0
  13. package/templates/commands/wbActOn/wbActOn_template.md +546 -0
  14. package/templates/commands/wbAudit/wbAudit_template.md +315 -0
  15. package/templates/commands/wbBroadcast/wbBroadcast_template.md +133 -0
  16. package/templates/commands/wbCheck/wbCheck_template.md +322 -0
  17. package/templates/commands/wbClean/wbClean_template.md +118 -0
  18. package/templates/commands/wbContext/wbContext_template.md +213 -0
  19. package/templates/commands/wbDebug/wbDebug_template.md +132 -0
  20. package/templates/commands/wbDeploy/wbDeploy_template.md +224 -0
  21. package/templates/commands/wbDoc/wbDoc_template.md +138 -0
  22. package/templates/commands/wbExplain/wbExplain_template.md +98 -0
  23. package/templates/commands/wbGit/wbGit_template.md +160 -0
  24. package/templates/commands/wbHelp/wbHelp_template.md +101 -0
  25. package/templates/commands/wbIdea/wbIdea_template.md +337 -0
  26. package/templates/commands/wbLicense/wbLicense_template.md +148 -0
  27. package/templates/commands/wbMonetize/wbMonetize_template.md +113 -0
  28. package/templates/commands/wbNext/wbNext_template.md +270 -0
  29. package/templates/commands/wbPlan/wbPlan_template.md +413 -0
  30. package/templates/commands/wbPublish/wbPublish_template.md +205 -0
  31. package/templates/commands/wbRefactor/wbRefactor_template.md +129 -0
  32. package/templates/commands/wbRelease/wbRelease_template.md +229 -0
  33. package/templates/commands/wbReview/wbReview_template.md +252 -0
  34. package/templates/commands/wbSecure/wbSecure_template.md +157 -0
  35. package/templates/commands/wbSetup/wbSetup_template.md +203 -0
  36. package/templates/commands/wbStandup/wbStandup_template.md +198 -0
  37. package/templates/commands/wbTest/wbTest_template.md +226 -0
  38. package/templates/commands/wbToWBC/wbToWBC_template.md +91 -0
  39. package/templates/commands/wbTrack/wbTrack_template.md +555 -0
  40. package/templates/commands/wbTranslate/wbTranslate_template.md +114 -0
  41. package/templates/commands/wbValid/wbValid_template.md +142 -0
  42. package/templates/commands/wbVision/wbVision_template.md +141 -0
  43. package/templates/commands/wbWork/wbWork_template.md +140 -0
  44. package/templates/commands/wb_commands_reference.claude.json +2305 -0
  45. package/templates/commands/wb_commands_reference.json +1109 -0
  46. package/templates/shortcuts/shortcuts.md +413 -0
  47. package/templates/shortcuts/usage-management-examples-shortcuts/budget_controllers_examples.md +96 -0
  48. package/templates/shortcuts/usage-management-examples-shortcuts/ultimate_shortcuts_examples.md +1531 -0
@@ -0,0 +1,1531 @@
1
+ # AI Shortcuts — Daily Use Training Examples
2
+ > Simulate these examples to train with every command in the ultra-lite system.
3
+ > Each section covers a command with 2-4 real-world scenarios.
4
+
5
+ ---
6
+
7
+ ## STYLE & LENGTH
8
+
9
+ ### /c — Short reply (2-3 sentences max)
10
+
11
+ **Scenario 1 — Quick definition**
12
+ ```
13
+ /c What is Docker?
14
+ ```
15
+ > Docker is a tool that packages your app and its dependencies into a container so it runs the same everywhere. Think of it as a lightweight VM without the overhead.
16
+
17
+ **Scenario 2 — Quick decision**
18
+ ```
19
+ /c Should I use REST or GraphQL for my MVP?
20
+ ```
21
+ > Use REST. It's simpler, faster to build, and easier to debug at MVP stage. Switch to GraphQL when your frontend data needs become complex.
22
+
23
+ ---
24
+
25
+ ### /d — Full detailed response
26
+
27
+ **Scenario 1 — Architecture decision**
28
+ ```
29
+ /d Explain the difference between monolith and microservices for a startup
30
+ ```
31
+ > [Full response covering tradeoffs, when to use each, migration path, real examples]
32
+
33
+ **Scenario 2 — Deep dive on a concept**
34
+ ```
35
+ /d How does JWT authentication work end to end?
36
+ ```
37
+ > [Full response: token structure, signing, verification, refresh tokens, security considerations]
38
+
39
+ ---
40
+
41
+ ### /wissem — Direct, startup-aware, Tunisia context
42
+
43
+ **Scenario 1 — Business advice**
44
+ ```
45
+ /wissem Should I incorporate in Tunisia or offshore for my SaaS?
46
+ ```
47
+ > Offshore (Estonia or Delaware) if you're targeting international clients — easier payments, better credibility with foreign investors. Tunisia if you're serving local market or need local banking. What's your target customer?
48
+
49
+ **Scenario 2 — Product decision**
50
+ ```
51
+ /wissem We have 3 features to build. How do I decide which first?
52
+ ```
53
+ > Build the one that unblocks revenue or retention first. Everything else is a distraction until you're making money.
54
+
55
+ ---
56
+
57
+ ### /direct — No hedging, no disclaimers
58
+
59
+ **Scenario 1 — Honest feedback**
60
+ ```
61
+ /direct Is my landing page copy good? "We leverage AI-powered synergies to optimize your workflow."
62
+ ```
63
+ > No. It's corporate jargon that says nothing. Rewrite it as what you actually do: "We automate [specific task] so your team saves [X hours] a week."
64
+
65
+ **Scenario 2 — Tech choice**
66
+ ```
67
+ /direct Should I learn React or Vue in 2026?
68
+ ```
69
+ > React. Bigger job market, more libraries, more tutorials. Vue is fine but React wins on career ROI.
70
+
71
+ ---
72
+
73
+ ### /vd — Exhaustive response
74
+
75
+ **Scenario 1 — Full audit**
76
+ ```
77
+ /vd Review all security considerations for a Node.js REST API
78
+ ```
79
+ > [Exhaustive: input validation, auth, rate limiting, CORS, headers, SQL injection, XSS, logging, secrets management, dependencies audit, HTTPS, etc.]
80
+
81
+ ---
82
+
83
+ ### /eli5 — Explain like I'm 5
84
+
85
+ **Scenario 1 — Technical concept**
86
+ ```
87
+ /eli5 What is an API?
88
+ ```
89
+ > Imagine a restaurant. You (the app) give your order to the waiter (the API). The waiter takes it to the kitchen (the server) and brings back your food (the data). You never go into the kitchen yourself.
90
+
91
+ **Scenario 2 — Business concept**
92
+ ```
93
+ /eli5 What is product-market fit?
94
+ ```
95
+ > It means people really want what you're selling. Like when everyone at school wants the same toy — the toy has found its market.
96
+
97
+ ---
98
+
99
+ ### /technical — Precise technical language
100
+
101
+ **Scenario 1 — Code review**
102
+ ```
103
+ /technical Explain what's wrong with using var in JavaScript
104
+ ```
105
+ > `var` is function-scoped and hoisted to the top of its scope, which can cause unintended variable leakage and temporal dead zone issues. Use `const` for immutable bindings and `let` for block-scoped mutable variables to enforce lexical scoping.
106
+
107
+ ---
108
+
109
+ ### /practical — Real-world focus
110
+
111
+ **Scenario 1 — Learning path**
112
+ ```
113
+ /practical How do I learn system design?
114
+ ```
115
+ > Start by reading real post-mortems (GitHub, Netflix, Cloudflare blogs). Then design systems you actually use — Uber, WhatsApp. Practice explaining tradeoffs out loud. Don't read theory without building something first.
116
+
117
+ ---
118
+
119
+ ### /casual — Relaxed tone
120
+
121
+ **Scenario 1 — Brainstorm**
122
+ ```
123
+ /casual Give me name ideas for a productivity app for devs
124
+ ```
125
+ > Oh this is fun — how about FlowState, DeepWork, or just something cheeky like NoBrainer? Or go meta: Unblocked, ZeroTab, or Focusd (no 'e' for that startup feel 😄)
126
+
127
+ ---
128
+
129
+ ### /neutral — Balanced view
130
+
131
+ **Scenario 1 — Controversial tech**
132
+ ```
133
+ /neutral Is AI going to replace developers?
134
+ ```
135
+ > Some argue AI will automate repetitive coding tasks and reduce demand for junior developers. Others point out that software complexity is growing faster than AI can keep up, creating more demand overall. The likely reality is a shift in the skill mix required, not wholesale replacement.
136
+
137
+ ---
138
+
139
+ ## OUTPUT FORMAT
140
+
141
+ ### /b — Bullet list
142
+
143
+ **Scenario 1 — Feature list**
144
+ ```
145
+ /b What should a good user onboarding flow include?
146
+ ```
147
+ > - Welcome email with clear next step
148
+ > - In-app checklist (3-5 items max)
149
+ > - Empty state guidance on first login
150
+ > - One "aha moment" trigger within first session
151
+ > - Follow-up email at day 3 if not activated
152
+
153
+ ---
154
+
155
+ ### /nb — Numbered list
156
+
157
+ **Scenario 1 — Process**
158
+ ```
159
+ /nb Steps to deploy a Node app to a VPS
160
+ ```
161
+ > 1. SSH into your server
162
+ > 2. Clone your repo
163
+ > 3. Run `npm install --production`
164
+ > 4. Set environment variables
165
+ > 5. Start with `pm2 start app.js`
166
+ > 6. Configure Nginx as reverse proxy
167
+ > 7. Set up SSL with Certbot
168
+
169
+ ---
170
+
171
+ ### /step — Step-by-step instructions
172
+
173
+ **Scenario 1 — Setup task**
174
+ ```
175
+ /step How do I set up GitHub Actions for CI on a Node project?
176
+ ```
177
+ > Step 1: Create `.github/workflows/ci.yml` in your repo
178
+ > Step 2: Define trigger — `on: [push, pull_request]`
179
+ > Step 3: Add a job with `runs-on: ubuntu-latest`
180
+ > Step 4: Add steps: checkout, setup-node, npm install, npm test
181
+ > Step 5: Push and check the Actions tab in GitHub
182
+
183
+ ---
184
+
185
+ ### /table — Structured table
186
+
187
+ **Scenario 1 — Comparison**
188
+ ```
189
+ /table Compare Vercel, Railway, and Render for deploying a Node API
190
+ ```
191
+ > | Platform | Free Tier | Cold Starts | Custom Domain | Best For |
192
+ > |---|---|---|---|---|
193
+ > | Vercel | Yes | Yes (serverless) | Yes | Frontend + edge |
194
+ > | Railway | $5 credit | No | Yes | Full-stack apps |
195
+ > | Render | Yes (limited) | Yes (free) | Yes | Side projects |
196
+
197
+ ---
198
+
199
+ ### /json — JSON output
200
+
201
+ **Scenario 1 — Data structure**
202
+ ```
203
+ /json Give me a user profile schema for a SaaS app
204
+ ```
205
+ > ```json
206
+ > {
207
+ > "id": "uuid",
208
+ > "email": "string",
209
+ > "name": "string",
210
+ > "plan": "free | pro | enterprise",
211
+ > "createdAt": "ISO8601",
212
+ > "settings": {
213
+ > "notifications": true,
214
+ > "language": "en"
215
+ > }
216
+ > }
217
+ > ```
218
+
219
+ ---
220
+
221
+ ### /code — Code block only
222
+
223
+ **Scenario 1 — Quick snippet**
224
+ ```
225
+ /code Node.js middleware to log request time
226
+ ```
227
+ > ```javascript
228
+ > app.use((req, res, next) => {
229
+ > const start = Date.now();
230
+ > res.on('finish', () => {
231
+ > console.log(`${req.method} ${req.url} - ${Date.now() - start}ms`);
232
+ > });
233
+ > next();
234
+ > });
235
+ > ```
236
+
237
+ ---
238
+
239
+ ### /proscons — Pros and cons
240
+
241
+ **Scenario 1 — Tech decision**
242
+ ```
243
+ /proscons Using MongoDB for a startup MVP
244
+ ```
245
+ > **Pros**
246
+ > - Flexible schema — great for evolving data models
247
+ > - Fast to start, no migrations early on
248
+ > - Good free tier on Atlas
249
+ >
250
+ > **Cons**
251
+ > - No joins — complex relations get messy fast
252
+ > - Easy to create performance problems with bad schemas
253
+ > - Not ideal if you need strong consistency
254
+
255
+ ---
256
+
257
+ ### /compare,a,b — Side-by-side comparison
258
+
259
+ **Scenario 1 — Framework choice**
260
+ ```
261
+ /compare,NextJS,NuxtJS
262
+ ```
263
+ > | | Next.js | Nuxt.js |
264
+ > |---|---|---|
265
+ > | Framework | React | Vue |
266
+ > | SSR | Yes | Yes |
267
+ > | Community | Larger | Smaller |
268
+ > | Learning curve | Medium | Easier |
269
+ > | Job market | Stronger | Good |
270
+ > | Best for | React devs, large teams | Vue devs, faster ramp-up |
271
+
272
+ ---
273
+
274
+ ### /final_only — Skip reasoning, show output only
275
+
276
+ **Scenario 1 — Get the result without explanation**
277
+ ```
278
+ /final_only Write a subject line for a cold email to a CTO about my dev tool
279
+ ```
280
+ > "Cut your team's code review time by 40% — worth 10 minutes?"
281
+
282
+ ---
283
+
284
+ ### /answer_only — One line
285
+
286
+ **Scenario 1 — Quick fact**
287
+ ```
288
+ /answer_only What HTTP status code means "not found"?
289
+ ```
290
+ > 404
291
+
292
+ ---
293
+
294
+ ### /no_explanation — Output only
295
+
296
+ **Scenario 1 — Clean code output**
297
+ ```
298
+ /no_explanation Write a Python function that checks if a string is a palindrome
299
+ ```
300
+ > ```python
301
+ > def is_palindrome(s):
302
+ > return s == s[::-1]
303
+ > ```
304
+
305
+ ---
306
+
307
+ ### /quick — Fastest useful answer
308
+
309
+ **Scenario 1 — Urgent fix**
310
+ ```
311
+ /quick How do I kill a process on port 3000 in Mac?
312
+ ```
313
+ > `lsof -ti:3000 | xargs kill`
314
+
315
+ ---
316
+
317
+ ## REASONING
318
+
319
+ ### /cot — Step-by-step reasoning
320
+
321
+ **Scenario 1 — Architecture decision**
322
+ ```
323
+ /cot Should I use a queue (like Bull) for sending emails in my app?
324
+ ```
325
+ > Let me think through this:
326
+ > 1. Email sending can fail — if done synchronously, it blocks the request
327
+ > 2. Users shouldn't wait for email delivery to get a response
328
+ > 3. If the email service is down, you need retry logic
329
+ > 4. At scale, bursts of signups would overwhelm a sync approach
330
+ > Conclusion: Yes, use a queue. Bull + Redis is the standard for Node. Even at small scale, it's the right pattern.
331
+
332
+ ---
333
+
334
+ ### /selfcritic — Answer → critique → improve
335
+
336
+ **Scenario 1 — Copy review**
337
+ ```
338
+ /selfcritic Write a tagline for a project management tool for dev teams
339
+ ```
340
+ > **Draft:** "Manage your projects smarter."
341
+ >
342
+ > **Critique:** Too generic. Every PM tool says this. No specificity, no benefit, no audience.
343
+ >
344
+ > **Improved:** "Ship faster. Less standup, more done." — speaks directly to dev teams' pain of too many meetings.
345
+
346
+ ---
347
+
348
+ ### /rootcause — Diagnose root cause
349
+
350
+ **Scenario 1 — Performance issue**
351
+ ```
352
+ /rootcause My API response times are increasing over time. What's the root cause?
353
+ ```
354
+ > Most likely root cause: unindexed database queries accumulating data over time.
355
+ > As your dataset grows, a query that was fast at 1K rows becomes slow at 100K rows without an index.
356
+ > Secondary suspects: memory leaks in long-running processes, N+1 query patterns, or connection pool exhaustion.
357
+ > Check: slow query logs first, then memory usage over time.
358
+
359
+ ---
360
+
361
+ ### /first_principles — Fundamentals first
362
+
363
+ **Scenario 1 — System design**
364
+ ```
365
+ /first_principles How should I design a notification system?
366
+ ```
367
+ > From first principles:
368
+ > - A notification is just a message triggered by an event
369
+ > - Events can be synchronous (user action) or async (scheduled)
370
+ > - Delivery channels are independent of the event (email, SMS, push)
371
+ > - Failures in delivery shouldn't block the core app
372
+ > Therefore: event → queue → worker → channel adapter. Each layer is independent. Start with this pattern and add complexity only when needed.
373
+
374
+ ---
375
+
376
+ ### /analogy — Explain with analogy
377
+
378
+ **Scenario 1 — Technical concept**
379
+ ```
380
+ /analogy Explain load balancing
381
+ ```
382
+ > Imagine a busy McDonald's with 4 cashiers. A load balancer is the person at the door who directs each customer to the cashier with the shortest queue. No single cashier gets overwhelmed, and customers get served faster overall.
383
+
384
+ ---
385
+
386
+ ### /improve — Make it better
387
+
388
+ **Scenario 1 — Improve existing code**
389
+ ```
390
+ /improve
391
+ [paste your code or text]
392
+ ```
393
+ > [Returns improved version with explanation of what changed and why]
394
+
395
+ ---
396
+
397
+ ### /simplify — Reduce complexity
398
+
399
+ **Scenario 1 — Simplify architecture**
400
+ ```
401
+ /simplify Our stack: React frontend, Node API, Redis cache, MongoDB, RabbitMQ, Elasticsearch. MVP with 50 users.
402
+ ```
403
+ > Drop RabbitMQ and Elasticsearch immediately. At 50 users you don't need a message queue or search engine. Use MongoDB's built-in text search. Add queues only when you have proven async load. Simpler stack = faster iteration = less ops burden.
404
+
405
+ ---
406
+
407
+ ### /expand — Add depth
408
+
409
+ **Scenario 1 — Expand a short answer**
410
+ ```
411
+ /expand You said "use indexes". Tell me more.
412
+ ```
413
+ > [Expanded: types of indexes, when to use compound vs single, how to identify missing indexes, explain plans, index tradeoffs on write performance, etc.]
414
+
415
+ ---
416
+
417
+ ### /rewrite — Full rewrite
418
+
419
+ **Scenario 1 — Rewrite bad copy**
420
+ ```
421
+ /rewrite "Our platform utilizes cutting-edge machine learning algorithms to provide actionable insights."
422
+ ```
423
+ > "We analyze your data and tell you exactly what to do next — no guesswork."
424
+
425
+ ---
426
+
427
+ ## CONTEXT & FILES
428
+
429
+ ### /ctx,name,load + /ctx,name,analyze
430
+
431
+ **Scenario 1 — Load and analyze backend**
432
+ ```
433
+ /ctx,backend,load
434
+ /ctx,backend,analyze
435
+ ```
436
+ > [Load]: Context "backend" loaded.
437
+ > [Analyze]: The backend uses Express with MongoDB. Auth is handled via JWT in middleware. Main concerns: no input validation on POST routes, database calls not wrapped in try/catch, no rate limiting on auth endpoints.
438
+
439
+ ---
440
+
441
+ ### /f,file,debug — Find bugs in a file
442
+
443
+ **Scenario 1 — Debug auth file**
444
+ ```
445
+ /f,auth.js,load
446
+ /f,auth.js,debug
447
+ ```
448
+ > Bugs found in auth.js:
449
+ > 1. Line 23: JWT secret hardcoded as string — should use `process.env.JWT_SECRET`
450
+ > 2. Line 47: No expiry set on token — tokens never expire
451
+ > 3. Line 61: Password compared with `==` instead of `bcrypt.compare()`
452
+
453
+ ---
454
+
455
+ ### /f,file,test — Generate tests
456
+
457
+ **Scenario 1 — Generate tests for a utility**
458
+ ```
459
+ /f,utils.js,load
460
+ /f,utils.js,test
461
+ ```
462
+ > [Returns full Jest test suite covering happy path, edge cases, and error cases]
463
+
464
+ ---
465
+
466
+ ### /f,file,refactor — Refactor for readability
467
+
468
+ **Scenario 1 — Clean up messy code**
469
+ ```
470
+ /f,userController.js,load
471
+ /f,userController.js,refactor
472
+ ```
473
+ > [Returns refactored version: extracted functions, consistent error handling, removed duplication, added clear variable names]
474
+
475
+ ---
476
+
477
+ ## EXECUTION
478
+
479
+ ### /plan_then_execute
480
+
481
+ **Scenario 1 — Build a feature**
482
+ ```
483
+ /wissem /goal,add_stripe_payments /plan_then_execute
484
+ ```
485
+ > Plan:
486
+ > 1. Install stripe SDK
487
+ > 2. Create /api/create-checkout-session endpoint
488
+ > 3. Add webhook handler for payment confirmation
489
+ > 4. Store subscription status on user model
490
+ > 5. Gate premium features behind status check
491
+ >
492
+ > Proceed? (yes/no)
493
+
494
+ ---
495
+
496
+ ### /dry_run — Simulate without executing
497
+
498
+ **Scenario 1 — Preview database migration**
499
+ ```
500
+ /dry_run Add a "status" field to all existing user records with default value "active"
501
+ ```
502
+ > Simulation:
503
+ > - Query: `db.users.updateMany({}, { $set: { status: "active" } })`
504
+ > - Affected documents: ~2,400 records
505
+ > - Risk: Low — additive change, no data loss
506
+ > - Rollback: `db.users.updateMany({}, { $unset: { status: "" } })`
507
+ > Run it? (yes/no)
508
+
509
+ ---
510
+
511
+ ### /goal,x + /goal_status
512
+
513
+ **Scenario 1 — Track progress**
514
+ ```
515
+ /goal,launch_beta_by_april
516
+ ```
517
+ > Goal set: launch_beta_by_april. I'll keep this in context throughout our session.
518
+
519
+ ```
520
+ /goal_status
521
+ ```
522
+ > Current goal: launch_beta_by_april
523
+ > Completed: auth, user dashboard, basic CRUD
524
+ > Remaining: payment integration, email notifications, staging deploy, QA
525
+ > Estimated: 60% done
526
+
527
+ ---
528
+
529
+ ## ROLES
530
+
531
+ ### /wissempro — Tunisian tech entrepreneur advisor
532
+
533
+ **Scenario 1 — Fundraising advice**
534
+ ```
535
+ /wissempro Should I raise a pre-seed round now or wait until I have revenue?
536
+ ```
537
+ > If you can get to first revenue in 3-6 months without raising, do it. Revenue changes your leverage completely — you're no longer begging, you're choosing. In Tunisia especially, bootstrap as far as you can before diluting. Raise only when you have a specific use of funds that multiplies revenue, not just to extend runway.
538
+
539
+ ---
540
+
541
+ ### /expert — World-class expert
542
+
543
+ **Scenario 1 — Security review**
544
+ ```
545
+ /expert Review the security of my JWT implementation
546
+ ```
547
+ > [Responds as a senior security engineer: checks algorithm (avoid none/HS256 weak keys), token storage (httpOnly cookies vs localStorage), expiry strategy, refresh token rotation, revocation approach, CSRF protection, etc.]
548
+
549
+ ---
550
+
551
+ ### /engineer — Pragmatic implementer
552
+
553
+ **Scenario 1 — Feature implementation**
554
+ ```
555
+ /engineer How should I implement search in my app?
556
+ ```
557
+ > At your scale: use your existing database's built-in full-text search first. PostgreSQL has `tsvector`, MongoDB has `$text`. Don't add Elasticsearch until you've proven the DB can't handle it. That's probably 100K+ documents with complex queries. Build the simple thing first.
558
+
559
+ ---
560
+
561
+ ### /critic — Honest evaluation
562
+
563
+ **Scenario 1 — Evaluate an idea**
564
+ ```
565
+ /critic My startup idea: an app that connects freelancers with clients in Tunisia
566
+ ```
567
+ > Hard pass on the current positioning. This is a crowded space globally (Upwork, Fiverr, Malt) and locally (there are already Tunisian copycats). Without a specific niche or unfair advantage, you'll spend all your money on supply/demand acquisition. What's your moat? If you can't answer that in one sentence, the idea needs more work.
568
+
569
+ ---
570
+
571
+ ### /pm — Scope and prioritize
572
+
573
+ **Scenario 1 — Scope an MVP**
574
+ ```
575
+ /pm We want to build a SaaS for invoice management. What's the MVP scope?
576
+ ```
577
+ > Cut it to: create invoice, send by email, mark as paid. That's it. No recurring invoices, no multi-currency, no client portal, no reminders, no dashboard. Those are v2. Ship the core loop first and validate that people will pay for it.
578
+
579
+ ---
580
+
581
+ ## PRIORITIZATION
582
+
583
+ ### /focus,top3
584
+
585
+ **Scenario 1 — Weekly planning**
586
+ ```
587
+ /wissempro /focus,top3 I have: fix login bug, write docs, add dark mode, improve onboarding, set up analytics, write blog post
588
+ ```
589
+ > Top 3:
590
+ > 1. Fix login bug — blocking users, revenue at risk
591
+ > 2. Improve onboarding — directly impacts activation rate
592
+ > 3. Set up analytics — you can't improve what you can't measure
593
+ >
594
+ > Everything else is noise until these are done.
595
+
596
+ ---
597
+
598
+ ### /quick_wins
599
+
600
+ **Scenario 1 — Find easy wins**
601
+ ```
602
+ /quick_wins What can I improve on my SaaS landing page this week?
603
+ ```
604
+ > - Add a real screenshot or demo GIF (2hrs, high conversion impact)
605
+ > - Add 2-3 customer logos or testimonials (1hr if you have them)
606
+ > - Make the CTA button bigger and change color to contrast (30min)
607
+ > - Add pricing on the page — hiding it kills trust (1hr)
608
+
609
+ ---
610
+
611
+ ### /effort — Effort estimate
612
+
613
+ **Scenario 1 — Estimate a feature**
614
+ ```
615
+ /effort Add Stripe subscription billing to an existing Node/React app
616
+ ```
617
+ > Medium-High effort. Estimate: 3-5 days for a solo dev.
618
+ > - Stripe SDK setup + webhook: 4hrs
619
+ > - Checkout flow + success/cancel pages: 4hrs
620
+ > - Subscription status on user model: 2hrs
621
+ > - Gating features behind plan: 3hrs
622
+ > - Testing + edge cases: 4hrs
623
+ > Risk: webhooks are the tricky part — budget extra time there.
624
+
625
+ ---
626
+
627
+ ## INTERACTION
628
+
629
+ ### /no_questions — Assume and proceed
630
+
631
+ **Scenario 1 — Code task**
632
+ ```
633
+ /no_questions /wissem Build me a REST endpoint to create a user
634
+ ```
635
+ > [Proceeds immediately with best assumptions: Node/Express, MongoDB, bcrypt password hashing, JWT response, input validation — no clarifying questions asked]
636
+
637
+ ---
638
+
639
+ ### /checklist — Actionable checklist
640
+
641
+ **Scenario 1 — Launch prep**
642
+ ```
643
+ /checklist What do I need before launching my SaaS publicly?
644
+ ```
645
+ > - [ ] Payment flow tested end-to-end
646
+ > - [ ] Error monitoring set up (Sentry)
647
+ > - [ ] Basic analytics in place
648
+ > - [ ] Terms of Service and Privacy Policy live
649
+ > - [ ] Transactional emails working (welcome, password reset)
650
+ > - [ ] Custom domain + SSL configured
651
+ > - [ ] Load tested with expected traffic
652
+ > - [ ] Support email or chat widget active
653
+ > - [ ] Backup strategy for database
654
+
655
+ ---
656
+
657
+ ### /fix — Fix and explain
658
+
659
+ **Scenario 1 — Fix a bug**
660
+ ```
661
+ /fix
662
+ async function getUser(id) {
663
+ const user = await User.findById(id)
664
+ return user.name
665
+ }
666
+ ```
667
+ > **Bug:** If `user` is null (ID not found), `user.name` throws `TypeError: Cannot read property 'name' of null`.
668
+ >
669
+ > **Fix:**
670
+ > ```javascript
671
+ > async function getUser(id) {
672
+ > const user = await User.findById(id);
673
+ > if (!user) throw new Error(`User ${id} not found`);
674
+ > return user.name;
675
+ > }
676
+ > ```
677
+
678
+ ---
679
+
680
+ ### /explain_error — Plain-language error explanation
681
+
682
+ **Scenario 1 — Node error**
683
+ ```
684
+ /explain_error ECONNREFUSED 127.0.0.1:5432
685
+ ```
686
+ > Your app is trying to connect to a PostgreSQL database on your local machine (port 5432) but nothing is running there. Either PostgreSQL isn't started, it's running on a different port, or your connection string is pointing to localhost when it should point to a remote host (like your Docker container or cloud DB).
687
+
688
+ ---
689
+
690
+ ## META
691
+
692
+ ### /help — Explain a shortcut
693
+
694
+ **Scenario 1**
695
+ ```
696
+ /help /selfcritic
697
+ ```
698
+ > `/selfcritic` makes me generate an answer, then immediately critique it for weaknesses, then produce an improved version. Use it when you want higher quality output, not just a first draft — especially useful for copy, strategy, or architecture decisions.
699
+
700
+ ---
701
+
702
+ ### /style — Show current active settings
703
+
704
+ **Scenario 1**
705
+ ```
706
+ /style
707
+ ```
708
+ > Current active settings:
709
+ > - Tone: /wissem (direct, startup-aware)
710
+ > - Length: /d (full detail)
711
+ > - Output: /final_only
712
+ > - Goal: launch_beta_by_april
713
+ > - Mode: /no_questions
714
+
715
+ ---
716
+
717
+ ### /business — ROI framing
718
+
719
+ **Scenario 1 — Tech decision**
720
+ ```
721
+ /business Should we rewrite our frontend in TypeScript?
722
+ ```
723
+ > Business case: TypeScript reduces bugs that reach production, which means less time spent on customer-reported issues and faster onboarding for new devs. Typical teams report 15-30% fewer runtime errors. Cost: 2-4 weeks of migration depending on codebase size. ROI breakeven: ~3-6 months. Recommend: yes, but incrementally — don't stop shipping to do it.
724
+
725
+ ---
726
+
727
+ ## POWER CHAINS — Full Workflow Examples
728
+
729
+ ### Chain 1 — Debug and fix a file
730
+ ```
731
+ /wissem /no_questions /engineer /f,auth.js,load /f,auth.js,debug /fix /final_only
732
+ ```
733
+ > [Loads file → finds bugs → fixes them → shows final fixed code only, no commentary]
734
+
735
+ ---
736
+
737
+ ### Chain 2 — Plan a feature from scratch
738
+ ```
739
+ /wissempro /goal,add_notifications /plan_then_execute /no_questions /step
740
+ ```
741
+ > [Sets goal → shows step-by-step plan → waits for approval → executes]
742
+
743
+ ---
744
+
745
+ ### Chain 3 — Write and improve copy
746
+ ```
747
+ /direct /selfcritic Write a cold email subject line for a dev tool targeting CTOs
748
+ ```
749
+ > Draft → Critique → Improved version
750
+
751
+ ---
752
+
753
+ ### Chain 4 — Quick startup decision
754
+ ```
755
+ /wissempro /c /direct Should I hire a junior dev or a senior dev first?
756
+ ```
757
+ > Hire senior first if you need architecture decisions made right. Hire junior if the foundation is solid and you need execution speed at lower cost. At pre-revenue: senior. Post-revenue with traction: junior to scale output.
758
+
759
+ ---
760
+
761
+ ### Chain 5 — Full session launcher
762
+ ```
763
+ ## AI Shortcuts — Ultra Trim
764
+
765
+ STYLE: /wissem=direct+startup+Tunisia | /direct=no hedging | /c=short | /d=full
766
+ OUTPUT: /b=bullets | /step=step-by-step | /final_only | /cot | /selfcritic
767
+ WORKFLOW: /no_questions | /goal,x | /execute | /plan_then_execute | /fix
768
+ Always remember active /goal until /reset.
769
+
770
+ ⚡ START: /wissem /no_questions /goal,launch_mvp /plan_then_execute /final_only
771
+ ```
772
+ > [Paste this → Claude loads all shortcuts + sets goal + starts planning immediately]
773
+
774
+ ---
775
+
776
+ *End of training examples. Practice 2-3 scenarios per day until chains feel natural.*
777
+
778
+
779
+ ## CONTEXT & FILE MANAGEMENT (Section 15)
780
+
781
+ ### /ctx,foldA,load — Load full content of a folder (paste immediately after). AI confirms with one-line summary.
782
+
783
+ **Scenario 1**
784
+ ```
785
+ /ctx,backend,load
786
+ [paste entire backend folder code here]
787
+ ```
788
+
789
+ > ✓ Context backend loaded — FastAPI app with auth, users, payments (8 files, 2,847 LOC, PostgreSQL + Redis).
790
+
791
+ ---
792
+
793
+ ### /ctx,foldA,set, — Define what the folder contains without pasting full code.
794
+
795
+ **Scenario 1**
796
+ ```
797
+ `/ctx,frontend,set,React + Tailwind dashboard with 5 pages and API hooks`
798
+ ```
799
+
800
+ > ✓ Context frontend set — React + Tailwind dashboard (5 pages, API hooks, no full code pasted).
801
+
802
+ ---
803
+
804
+ ### /ctx,foldA,scope, — Narrow focus inside an already loaded folder.
805
+
806
+ **Scenario 1**
807
+ ```
808
+ `/ctx,backend,scope,auth_module`
809
+ ```
810
+
811
+ > Scope narrowed to auth_module only (JWT, refresh tokens, rate limiting).
812
+
813
+ ---
814
+
815
+ ### /ctx,foldA,reset — Clear a context slot.
816
+
817
+ **Scenario 1**
818
+ ```
819
+ `/ctx,backend,reset`
820
+ ```
821
+
822
+ > ✓ Context backend cleared.
823
+
824
+ ---
825
+
826
+ ### /ctx,list — Show all active context slots.
827
+
828
+ **Scenario 1**
829
+ ```
830
+ `/ctx,list`
831
+ ```
832
+
833
+ > Active contexts:
834
+ > • backend (loaded)
835
+ > • frontend (set)
836
+ > • auth.py (loaded)
837
+
838
+ ---
839
+
840
+ ### /f,file.py,load — Load single file content.
841
+
842
+ **Scenario 1**
843
+ ```
844
+ /f,auth.py,load
845
+ [paste full auth.py here]
846
+ ```
847
+
848
+ > ✓ File auth.py loaded — 412 LOC, FastAPI auth endpoints + dependency injection.
849
+
850
+ ---
851
+
852
+ ### /f,file.py,set, — Define file without full paste.
853
+
854
+ **Scenario 1**
855
+ ```
856
+ `/f,utils.py,set,helper functions for validation and logging`
857
+ ```
858
+
859
+ > ✓ File utils.py set.
860
+ >
861
+
862
+ ---
863
+
864
+ ### /ctx,foldA,overview — High-level one-paragraph summary.
865
+
866
+ **Scenario 1**
867
+ ```
868
+ /ctx,foldA,overview
869
+ ```
870
+
871
+ > Backend folder contains a production-ready FastAPI service with clean architecture (routers, services, repositories). Total 8 modules, main entrypoint main.py.
872
+
873
+ ---
874
+
875
+ ### /ctx,foldA,analyze — Deep structural + logic analysis.
876
+
877
+ **Scenario 1**
878
+ ```
879
+ /ctx,foldA,analyze
880
+ ```
881
+
882
+ > Structural analysis: 3 layers (routers → services → repositories). Circular dependency detected between payments and users. Logic gaps in error handling for 429 rate limits.
883
+
884
+ ---
885
+
886
+ ### /ctx,foldA,document — Generate full documentation.
887
+
888
+ **Scenario 1**
889
+ ```
890
+ /ctx,foldA,document
891
+ ```
892
+
893
+ > # Backend Documentation
894
+ > ## Endpoints
895
+ > • POST /auth/login → returns JWT
896
+ > ...
897
+
898
+ ---
899
+
900
+ ### /ctx,foldA,improve — Suggest quality improvements.
901
+
902
+ **Scenario 1**
903
+ ```
904
+ /ctx,foldA,improve
905
+ ```
906
+
907
+ > Recommendations: 1. Add pydantic v2 models 2. Replace print() with structured logging 3. Add OpenAPI tags.
908
+
909
+ ---
910
+
911
+ ### /ctx,foldA,refactor — Refactor for clarity/maintainability.
912
+
913
+ **Scenario 1**
914
+ ```
915
+ /ctx,foldA,refactor
916
+ ```
917
+
918
+ > Refactored structure proposed + diff view of all files.
919
+
920
+ ---
921
+
922
+ ### /ctx,foldA,debug — Find bugs and issues.
923
+
924
+ **Scenario 1**
925
+ ```
926
+ /ctx,foldA,debug
927
+ ```
928
+
929
+ > Critical: unhandled exception in payment webhook (line 142). Warning: SQL injection risk in raw query (users.py:89).
930
+
931
+ ---
932
+
933
+ ### /ctx,foldA,secure — Security audit.
934
+
935
+ **Scenario 1**
936
+ ```
937
+ /ctx,foldA,secure
938
+ ```
939
+
940
+ > Vulnerabilities found: 1. Missing rate-limit on login 2. Exposed stack traces in prod 3. Hard-coded secrets.
941
+
942
+ ---
943
+
944
+ ### /ctx,foldA,optimize — Performance improvements.
945
+
946
+ **Scenario 1**
947
+ ```
948
+ /ctx,foldA,optimize
949
+ ```
950
+
951
+ > Optimizations: Add Redis caching to user service → -65% DB load. Use async SQLAlchemy. Index on payments.created_at.
952
+
953
+ ---
954
+
955
+ ### /ctx,foldA,test — Generate test cases.
956
+
957
+ **Scenario 1**
958
+ ```
959
+ /ctx,foldA,test
960
+ ```
961
+
962
+ > Generated 28 pytest cases (unit + integration) for the entire folder.
963
+
964
+ ---
965
+
966
+ ### /ctx,foldA,map — Visual text-based map of structure.
967
+
968
+ **Scenario 1**
969
+ ```
970
+ /ctx,foldA,map
971
+ ```
972
+
973
+ ```
974
+ > backend/
975
+ > ├── routers/
976
+ > ├── services/
977
+ > ├── repositories/
978
+ > └── main.py (entry)
979
+ ```
980
+
981
+ ---
982
+
983
+ ### /ctx,foldA,explain — Plain-language explanation.
984
+
985
+ **Scenario 1**
986
+ ```
987
+ /ctx,foldA,explain
988
+ ```
989
+
990
+ > This folder is the complete backend API that handles user authentication, payments, and profile management.
991
+
992
+ ---
993
+
994
+ ### /ctx,foldA,dependencies — List internal + external dependencies.
995
+
996
+ **Scenario 1**
997
+ ```
998
+ /ctx,foldA,dependencies
999
+ ```
1000
+
1001
+ > Internal: auth → users → payments
1002
+ > External: fastapi, sqlalchemy, redis, stripe.
1003
+
1004
+ ---
1005
+
1006
+ ### /ctx,foldA,entrypoints — Identify all entry points.
1007
+
1008
+ **Scenario 1**
1009
+ ```
1010
+ /ctx,foldA,entrypoints
1011
+ ```
1012
+
1013
+ > • main.py:uvicorn
1014
+ > • routers/auth.py
1015
+ > • webhooks/stripe.py
1016
+
1017
+ ---
1018
+
1019
+ ### /ctx,foldA,architecture — Describe architectural pattern.
1020
+
1021
+ **Scenario 1**
1022
+ ```
1023
+ /ctx,foldA,architecture
1024
+ ```
1025
+
1026
+ > Clean Architecture + Repository pattern. Layers are strictly separated.
1027
+
1028
+ ---
1029
+
1030
+ ### /ctx,foldA,dataflow — Trace data flow.
1031
+
1032
+ **Scenario 1**
1033
+ ```
1034
+ /ctx,foldA,dataflow
1035
+ ```
1036
+
1037
+ > Request → Router → Service → Repository → DB → Response (with caching layer).
1038
+
1039
+ ---
1040
+
1041
+ ### /ctx,foldA,public_docs — Public-facing documentation.
1042
+
1043
+ **Scenario 1**
1044
+ ```
1045
+ /ctx,foldA,public_docs
1046
+ ```
1047
+
1048
+ > Ready-to-publish README + Swagger description.
1049
+
1050
+ ---
1051
+
1052
+ ### /ctx,foldA,expose — Show exposed vs internal.
1053
+
1054
+ **Scenario 1**
1055
+ ```
1056
+ /ctx,foldA,expose
1057
+ ```
1058
+
1059
+ > Public API: 12 endpoints. Internal only: 8 repository classes.
1060
+
1061
+ ---
1062
+
1063
+ ### /ctx,foldA,summary — Fast condensed overview.
1064
+
1065
+ **Scenario 1**
1066
+ ```
1067
+ /ctx,foldA,summary
1068
+ ```
1069
+
1070
+ > Backend: FastAPI + PostgreSQL, 8 modules, auth + payments core.
1071
+
1072
+ ---
1073
+
1074
+ ### /ctx,foldA,todos — Extract all TODO/FIXME/HACK.
1075
+
1076
+ **Scenario 1**
1077
+ ```
1078
+ /ctx,foldA,todos
1079
+ ```
1080
+
1081
+ > • TODO: implement email verification (auth.py:67)
1082
+ > • HACK: temporary stripe test key (payments.py:22)
1083
+
1084
+ ---
1085
+
1086
+ ### /ctx,foldA,lint — Flag style/quality issues.
1087
+
1088
+ **Scenario 1**
1089
+ ```
1090
+ /ctx,foldA,lint
1091
+ ```
1092
+
1093
+ > 23 issues: 12 missing type hints, 7 long functions, 4 unused imports.
1094
+
1095
+ ---
1096
+
1097
+ ### /ctx,foldA,changelog — Summarize changes (when diffs provided).
1098
+
1099
+ **Scenario 1**
1100
+ ```
1101
+ /ctx,foldA,changelog
1102
+ ```
1103
+
1104
+ > Changelog summary: Added webhook handler, refactored auth service, bumped FastAPI 0.110 → 0.115.
1105
+ >
1106
+
1107
+ ---
1108
+
1109
+ ### /ctx_multi,foldA,foldB,compare —
1110
+
1111
+ **Scenario 1**
1112
+ ```
1113
+ `/ctx_multi,backend,frontend,compare`
1114
+ ```
1115
+
1116
+ > Comparison table: tech stack, file count, coupling level, etc.
1117
+
1118
+ ---
1119
+
1120
+ ### /ctx_merge,foldA,foldB —
1121
+
1122
+ **Scenario 1**
1123
+ ```
1124
+ `/ctx_merge,backend,frontend`
1125
+ ```
1126
+
1127
+ > Merged context active — full-stack view now available.
1128
+
1129
+ ---
1130
+
1131
+ ### /ctx_diff,foldA,foldB —
1132
+
1133
+ **Scenario 1**
1134
+ ```
1135
+ `/ctx_diff,backend_v1,backend_v2`
1136
+ ```
1137
+
1138
+ > Diff summary: +3 files, -124 LOC, new payment module added.
1139
+ >
1140
+
1141
+ ---
1142
+
1143
+ ### /f,file.py,explain —
1144
+
1145
+ **Scenario 1**
1146
+ ```
1147
+ /f,file.py,explain
1148
+ ```
1149
+
1150
+ > auth.py implements JWT login, refresh, and dependency injection for protected routes.
1151
+
1152
+ ---
1153
+
1154
+ ### /f,file.py,refactor —
1155
+
1156
+ **Scenario 1**
1157
+ ```
1158
+ /f,file.py,refactor
1159
+ ```
1160
+
1161
+ > Refactored version with better separation + diff.
1162
+
1163
+ ---
1164
+
1165
+ ### /f,file.py,debug —
1166
+
1167
+ **Scenario 1**
1168
+ ```
1169
+ /f,file.py,debug
1170
+ ```
1171
+
1172
+ > Bug found: missing await on async DB call (line 134).
1173
+
1174
+ ---
1175
+
1176
+ ### /f,file.py,document —
1177
+
1178
+ **Scenario 1**
1179
+ ```
1180
+ /f,file.py,document
1181
+ ```
1182
+
1183
+ > Added full docstrings + module-level docs.
1184
+
1185
+ ---
1186
+
1187
+ ### /f,file.py,optimize —
1188
+
1189
+ **Scenario 1**
1190
+ ```
1191
+ /f,file.py,optimize
1192
+ ```
1193
+
1194
+ > Performance improved: cached query, reduced DB calls by 40%.
1195
+
1196
+ ---
1197
+
1198
+ ### /f,file.py,test —
1199
+
1200
+ **Scenario 1**
1201
+ ```
1202
+ /f,file.py,test
1203
+ ```
1204
+
1205
+ > Generated 12 pytest functions covering all paths.
1206
+
1207
+ ---
1208
+
1209
+ ### /f,file.py,review —
1210
+
1211
+ **Scenario 1**
1212
+ ```
1213
+ /f,file.py,review
1214
+ ```
1215
+
1216
+ > Code review: 9/10. Suggestions: add logging, improve error messages.
1217
+
1218
+ ---
1219
+
1220
+ ### /f,file.py,convert,lang —
1221
+
1222
+ **Scenario 1**
1223
+ ```
1224
+ `/f,auth.py,convert,typescript`
1225
+ ```
1226
+
1227
+ > Full conversion to TypeScript + NestJS style.
1228
+
1229
+ ---
1230
+
1231
+ ### /f,file.py,improve —
1232
+
1233
+ **Scenario 1**
1234
+ ```
1235
+ /f,file.py,improve
1236
+ ```
1237
+
1238
+ > General improvements applied (type hints, error handling, tests).
1239
+
1240
+ ---
1241
+
1242
+ ### /f,file.py,secure —
1243
+
1244
+ **Scenario 1**
1245
+ ```
1246
+ /f,file.py,secure
1247
+ ```
1248
+
1249
+ > Security audit passed except one minor issue (fixed version attached).
1250
+
1251
+ ---
1252
+
1253
+ ### /f,file.py,summary —
1254
+
1255
+ **Scenario 1**
1256
+ ```
1257
+ /f,file.py,summary
1258
+ ```
1259
+
1260
+ > One-paragraph summary of the file.
1261
+
1262
+ ---
1263
+
1264
+ ### /f,file.py,trace, —
1265
+
1266
+ **Scenario 1**
1267
+ ```
1268
+ `/f,auth.py,trace,login_function`
1269
+ ```
1270
+
1271
+ > Step-by-step execution trace of login_function.
1272
+
1273
+ ---
1274
+
1275
+ ### /f,file.py,mock —
1276
+
1277
+ **Scenario 1**
1278
+ ```
1279
+ /f,file.py,mock
1280
+ ```
1281
+
1282
+ > Complete mock/stub file generated.
1283
+
1284
+ ---
1285
+
1286
+ ### /f,file.py,diff,file2.py —
1287
+
1288
+ **Scenario 1**
1289
+ ```
1290
+ /f,file.py,diff,file2.py
1291
+ ```
1292
+
1293
+ > Side-by-side diff between the two files.
1294
+ >
1295
+
1296
+ ---
1297
+
1298
+ ### /p,projectX,health —
1299
+
1300
+ **Scenario 1**
1301
+ ```
1302
+ /p,projectX,health
1303
+ ```
1304
+
1305
+ > Project health: 92/100 (strong architecture, 3 minor tech debt items).
1306
+
1307
+ ---
1308
+
1309
+ ### /p,projectX,roadmap —
1310
+
1311
+ **Scenario 1**
1312
+ ```
1313
+ /p,projectX,roadmap
1314
+ ```
1315
+
1316
+ > Suggested 3-month roadmap with priorities.
1317
+
1318
+ ---
1319
+
1320
+ ### /p,projectX,risks —
1321
+
1322
+ **Scenario 1**
1323
+ ```
1324
+ /p,projectX,risks
1325
+ ```
1326
+
1327
+ > Technical risks: DB scaling at 10k users. Business risks: payment provider dependency.
1328
+
1329
+ ---
1330
+
1331
+ ### /p,projectX,scalability —
1332
+
1333
+ **Scenario 1**
1334
+ ```
1335
+ /p,projectX,scalability
1336
+ ```
1337
+
1338
+ > Scalability assessment: ready for 50k users with current design.
1339
+
1340
+ ---
1341
+
1342
+ ### /p,projectX,readiness —
1343
+
1344
+ **Scenario 1**
1345
+ ```
1346
+ /p,projectX,readiness
1347
+ ```
1348
+
1349
+ > Production readiness: 85% — needs observability + CI/CD finalised.
1350
+
1351
+ ---
1352
+
1353
+ ### /p,projectX,architecture —
1354
+
1355
+ **Scenario 1**
1356
+ ```
1357
+ /p,projectX,architecture
1358
+ ```
1359
+
1360
+ > Full architectural review + diagrams (text).
1361
+ >
1362
+
1363
+ ---
1364
+
1365
+ ### /save — Format current output for saving.
1366
+
1367
+ **Scenario 1**
1368
+ ```
1369
+ /save
1370
+ ```
1371
+
1372
+
1373
+
1374
+ ---
1375
+
1376
+ ### /save_as,file.md —
1377
+
1378
+ **Scenario 1**
1379
+ ```
1380
+ /save_as,file.md
1381
+ ```
1382
+
1383
+
1384
+
1385
+ ---
1386
+
1387
+ ### /save_to,foldA,file.md —
1388
+
1389
+ **Scenario 1**
1390
+ ```
1391
+ /save_to,foldA,file.md
1392
+ ```
1393
+
1394
+
1395
+
1396
+ ---
1397
+
1398
+ ### /export,markdown —
1399
+
1400
+ **Scenario 1**
1401
+ ```
1402
+ /export,markdown
1403
+ ```
1404
+
1405
+
1406
+
1407
+ ---
1408
+
1409
+ ### /export_to,foldA,file.md —
1410
+
1411
+ **Scenario 1**
1412
+ ```
1413
+ /export_to,foldA,file.md
1414
+ ```
1415
+
1416
+ >
1417
+ > ---
1418
+ >
1419
+ > ### How the full version integrates (all other commands covered via real chains)
1420
+ >
1421
+ > Here are ready-to-copy chains that use **every category** from the full v4.3:
1422
+ >
1423
+ > 1. **Style + Tone + Format + Reasoning**
1424
+ > `/d /wissem /engineer /cot /table /ctx,backend,load [paste] /ctx,backend,analyze /final_only`
1425
+ >
1426
+ > 2. **Deep file work**
1427
+ > `/f,auth.py,load [paste] /f,auth.py,debug /f,auth.py,secure /f,auth.py,refactor /selfcritic /final_only`
1428
+ >
1429
+ > 3. **Project level**
1430
+ > `/p,projectX,load [paste all] /p,projectX,health /p,projectX,roadmap /p,projectX,risks /swot /goal,launch_mvp /execute`
1431
+ >
1432
+ > 4. **Multi-context + compare**
1433
+ > `/ctx_multi,backend,frontend,compare /ctx_diff,backend,backend_v2 /vd /proscons`
1434
+ >
1435
+ > 5. **Creative + goal driven**
1436
+ > `/creative /goal,build_api /ctx,api,load [paste] /ctx,api,optimize /brainstorm,5 /ep,optimize`
1437
+ >
1438
+ > Just paste the universal prefix once at the start of any new chat, then use any of the above chains + the file commands in this guide.
1439
+ >
1440
+ > You now have **complete, exhaustive, copy-paste-ready examples** for every single command in the full version. Save the file and you’re done.
1441
+ >
1442
+ > Ready when you are — drop your first context and let’s go.
1443
+
1444
+ ---
1445
+
1446
+
1447
+
1448
+ ## BUDGET & TOKEN MANAGEMENT (Section 22)
1449
+
1450
+
1451
+
1452
+ ---
1453
+
1454
+
1455
+ **Use Case:** You are about to run a massive refactor across 15 files and want to avoid an accidental $5.00 API charge.
1456
+
1457
+ ```
1458
+ /quote Refactor the entire authentication module in src/auth/ to use JWT instead of sessions.```
1459
+ > **Token Estimation:**
1460
+ > - Input context: ~45,000 tokens
1461
+ > - Expected output: ~8,000 tokens across 5 files
1462
+ > - **Estimated Cost:** $0.85
1463
+ >
1464
+ > *Type "Approve" to proceed with the execution.*
1465
+
1466
+ ---
1467
+
1468
+
1469
+ **Use Case:** You want the AI to do a task, but only if it's cheap. If it's an expensive task, you want to do it yourself using a guided prompt.
1470
+
1471
+ ```
1472
+ /le,10 Generate a comprehensive test suite for the dashboard components.`
1473
+
1474
+ **AI Behavior (Scenario A - Under 10%):**
1475
+ > *Task requires ~5% of token limit. Executing...*
1476
+ > [Outputs the test suite code immediately]
1477
+
1478
+ **AI Behavior (Scenario B - Over 10%):**
1479
+ > *Task rejected. Estimated token usage (35%) exceeds your 10% limit.*
1480
+ >
1481
+ > **Do-It-Yourself Execution Prompt:**
1482
+ > "Please copy and paste the following prompt into your terminal or a cheaper AI session to generate the tests file-by-file:
1483
+ > *'Read `src/dashboard/Chart.vue` and generate Vitest unit tests covering mounting and data-fetching. Save it to `tests/Chart.spec.js`.'* (Repeat for each component)."
1484
+
1485
+ ---
1486
+
1487
+
1488
+ **Use Case:** You are running an ongoing task and want to pause periodically to review the cost and approve the next chunk.
1489
+
1490
+ ```
1491
+ /usage,15 Migrate all old Vue 2 syntax to Vue 3 Composition API in the components folder.```
1492
+ > *Executing chunk 1...*
1493
+ > [Migrates 3 files]
1494
+ >
1495
+ > **⏸️ Token Limit Paused**
1496
+ > I have reached your 15% token usage threshold (Estimated cost spent: $0.45).
1497
+ > **Remaining work:** 7 files left.
1498
+ > **Estimated remaining cost:** $1.10 (approx. 35% more token usage).
1499
+ >
1500
+ > *Do you approve continuing for a custom chunk of 20%?*
1501
+
1502
+ ---
1503
+
1504
+
1505
+ **Use Case:** You've been chatting with the AI for 2 hours. The context window is huge, and every single prompt is costing $0.10 just in input tokens. You only need a small CSS fix now.
1506
+
1507
+ ```
1508
+ /diet Fix the margin on the login button in Header.vue.```
1509
+ > *Flushing conversational history and unrelated files...*
1510
+ > *Retaining only `Header.vue` lines 45-60.*
1511
+ >
1512
+ > Here is the fixed CSS class for the login button:
1513
+ > ```css
1514
+ > .login-btn { margin-left: 1rem; }
1515
+ > ```
1516
+ > *(Input tokens reduced from 80k to 400 tokens).*
1517
+
1518
+ ---
1519
+
1520
+
1521
+ **Use Case:** You are in an expensive Opus/Pro session, but you need a simple regex or a basic boilerplate script that a cheap model like Haiku/Flash could do for pennies.
1522
+
1523
+ ```
1524
+ /delegate Write a regex to validate Tunisian phone numbers (+216).```
1525
+ > *Task evaluated as trivial. Executing this on the current premium model is a waste of budget.*
1526
+ >
1527
+ > **Action:** Copy and paste the prompt below into a Gemini Flash or Claude Haiku session:
1528
+ > > *"Write a javascript regex that strictly matches Tunisian phone numbers starting with +216 followed by 8 digits. Provide 3 test cases."*
1529
+
1530
+ ---
1531
+