speccrew 0.1.12 → 0.2.1
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/.speccrew/agents/speccrew-feature-designer.md +120 -0
- package/.speccrew/agents/speccrew-product-manager.md +54 -0
- package/.speccrew/agents/speccrew-system-designer.md +150 -14
- package/.speccrew/agents/speccrew-system-developer.md +309 -37
- package/.speccrew/agents/speccrew-task-worker.md +43 -0
- package/.speccrew/agents/speccrew-team-leader.md +108 -11
- package/.speccrew/agents/speccrew-test-manager.md +278 -0
- package/.speccrew/skills/speccrew-dev-backend/SKILL.md +44 -0
- package/.speccrew/skills/speccrew-dev-desktop/SKILL.md +44 -0
- package/.speccrew/skills/speccrew-dev-frontend/SKILL.md +44 -0
- package/.speccrew/skills/speccrew-dev-mobile/SKILL.md +44 -0
- package/.speccrew/skills/speccrew-fd-api-contract/SKILL.md +70 -0
- package/.speccrew/skills/speccrew-fd-feature-design/SKILL.md +158 -0
- package/.speccrew/skills/speccrew-knowledge-bizs-api-analyze/SKILL.md +59 -29
- package/.speccrew/skills/speccrew-knowledge-bizs-dispatch/SKILL.md +37 -15
- package/.speccrew/skills/speccrew-knowledge-bizs-dispatch/STATUS-FORMATS.md +29 -4
- package/.speccrew/skills/speccrew-knowledge-bizs-dispatch/scripts/process-batch-results.js +71 -4
- package/.speccrew/skills/speccrew-knowledge-bizs-ui-analyze/SKILL.md +60 -30
- package/.speccrew/skills/speccrew-pm-requirement-analysis/SKILL.md +65 -0
- package/.speccrew/skills/speccrew-sd-backend/SKILL.md +38 -0
- package/.speccrew/skills/speccrew-sd-desktop/SKILL.md +38 -0
- package/.speccrew/skills/speccrew-sd-frontend/SKILL.md +38 -0
- package/.speccrew/skills/speccrew-sd-mobile/SKILL.md +38 -0
- package/.speccrew/skills/speccrew-test-case-design/SKILL.md +33 -0
- package/.speccrew/skills/speccrew-test-code-gen/SKILL.md +34 -0
- package/.speccrew/skills/speccrew-test-execute/SKILL.md +34 -0
- package/README.ar.md +70 -3
- package/README.bn.md +52 -0
- package/README.bs.md +70 -3
- package/README.da.md +70 -3
- package/README.de.md +70 -3
- package/README.el.md +52 -0
- package/README.en.md +69 -2
- package/README.es.md +70 -3
- package/README.fr.md +70 -3
- package/README.it.md +70 -3
- package/README.ja.md +70 -3
- package/README.ko.md +70 -3
- package/README.md +69 -2
- package/README.no.md +70 -3
- package/README.pl.md +70 -3
- package/README.pt-BR.md +70 -3
- package/README.ru.md +70 -3
- package/README.th.md +69 -2
- package/README.tr.md +69 -2
- package/README.uk.md +69 -2
- package/README.vi.md +52 -0
- package/README.zh-TW.md +70 -3
- package/docs/GETTING-STARTED.ar.md +78 -4
- package/docs/GETTING-STARTED.bn.md +78 -4
- package/docs/GETTING-STARTED.bs.md +78 -4
- package/docs/GETTING-STARTED.da.md +78 -4
- package/docs/GETTING-STARTED.de.md +78 -4
- package/docs/GETTING-STARTED.el.md +78 -4
- package/docs/GETTING-STARTED.en.md +78 -4
- package/docs/GETTING-STARTED.es.md +78 -4
- package/docs/GETTING-STARTED.fr.md +78 -4
- package/docs/GETTING-STARTED.it.md +78 -4
- package/docs/GETTING-STARTED.ja.md +79 -5
- package/docs/GETTING-STARTED.ko.md +79 -5
- package/docs/GETTING-STARTED.md +78 -4
- package/docs/GETTING-STARTED.no.md +78 -4
- package/docs/GETTING-STARTED.pl.md +78 -4
- package/docs/GETTING-STARTED.pt-BR.md +78 -4
- package/docs/GETTING-STARTED.ru.md +78 -4
- package/docs/GETTING-STARTED.th.md +79 -5
- package/docs/GETTING-STARTED.tr.md +78 -4
- package/docs/GETTING-STARTED.uk.md +78 -4
- package/docs/GETTING-STARTED.vi.md +79 -5
- package/docs/GETTING-STARTED.zh-TW.md +79 -5
- package/package.json +1 -1
- package/.speccrew/skills/speccrew-create-agents/SKILL.md +0 -98
- package/.speccrew/skills/speccrew-create-agents/templates/agents/designer-agent.md +0 -54
- package/.speccrew/skills/speccrew-create-agents/templates/agents/dev-agent.md +0 -79
- package/.speccrew/skills/speccrew-create-agents/templates/agents/test-agent.md +0 -80
- package/.speccrew/skills/speccrew-project-diagnosis/SKILL.md +0 -233
- package/.speccrew/skills/speccrew-project-diagnosis/templates/DIAGNOSIS-REPORT-TEMPLATE.md +0 -202
- package/.speccrew/skills/speccrew-workflow-diagnose/SKILL.md +0 -155
- package/workspace-template/docs/solutions/Agent/346/212/200/350/203/275/345/256/232/344/271/211+/351/234/200/346/261/202/346/226/207/346/241/243+UML/344/275/277/347/224/250/346/250/241/346/235/277/357/274/210ISA-95/345/205/255/346/256/265/345/274/217/350/236/215/345/220/210/347/211/210/357/274/211.md +0 -586
- package/workspace-template/docs/solutions/harness.md +0 -410
package/README.bn.md
CHANGED
|
@@ -41,6 +41,58 @@ SpecCrew একটি এম্বেডেড ভার্চুয়াল AI
|
|
|
41
41
|
|
|
42
42
|
---
|
|
43
43
|
|
|
44
|
+
## ✨ মূল বৈশিষ্ট্য
|
|
45
|
+
|
|
46
|
+
### 🏭 ভার্চুয়াল সফ্টওয়্যার টিম
|
|
47
|
+
এক ক্লিকে **7টি পেশাদার Agent ভূমিকা** + **30+টি Skill ওয়ার্কফ্লো** তৈরি, একটি সম্পূর্ণ ভার্চুয়াল সফ্টওয়্যার টিম গঠন:
|
|
48
|
+
- **Team Leader** - গ্লোবাল পরিকল্পনা এবং ইটারেশন ব্যবস্থাপনা
|
|
49
|
+
- **Product Manager** - প্রয়োজনীয়তা বিশ্লেষণ এবং PRD আউটপুট
|
|
50
|
+
- **Feature Designer** - বৈশিষ্ট্য ডিজাইন + API চুক্তি
|
|
51
|
+
- **System Designer** - Frontend/Backend/Mobile/Desktop সিস্টেম ডিজাইন
|
|
52
|
+
- **System Developer** - মাল্টিপ্ল্যাটফর্ম সমান্তরাল উন্নয়ন
|
|
53
|
+
- **Test Manager** - তিন-পর্যায়ের পরীক্ষা সমন্বয়
|
|
54
|
+
- **Task Worker** - সমান্তরাল উপ-কার্য সম্পাদন
|
|
55
|
+
|
|
56
|
+
### 📐 ISA-95 ছয়-পর্যায়ের মডেলিং
|
|
57
|
+
আন্তর্জাতিক **ISA-95** মডেলিং পদ্ধতির উপর ভিত্তি করে, ব্যবসায়িক প্রয়োজনীয়তা থেকে সফ্টওয়্যার সিস্টেমে রূপান্তরের মানকায়ন:
|
|
58
|
+
```
|
|
59
|
+
Domain Descriptions → Functions in Domains → Functions of Interest
|
|
60
|
+
↓ ↓ ↓
|
|
61
|
+
Information Flows → Categories of Information → Information Descriptions
|
|
62
|
+
```
|
|
63
|
+
- প্রতিটি পর্যায় নির্দিষ্ট UML ডায়াগ্রামের সাথে সামঞ্জস্যপূর্ণ (use case, sequence, class)
|
|
64
|
+
- ব্যবসায়িক প্রয়োজনীয়তা "ধাপে ধাপে পরিশোধিত" হয়, তথ্য ক্ষতি ছাড়াই
|
|
65
|
+
- আউটপুটগুলি উন্নয়নের জন্য সরাসরি ব্যবহারযোগ্য
|
|
66
|
+
|
|
67
|
+
### 📚 জ্ঞান ভিত্তি সিস্টেম
|
|
68
|
+
তিন-স্তরের জ্ঞান ভিত্তি আর্কিটেকচার নিশ্চিত করে যে AI সর্বদা "একক সত্যের উৎস" এর উপর ভিত্তি করে কাজ করে:
|
|
69
|
+
|
|
70
|
+
| স্তর | ডিরেক্টরি | বিষয়বস্তু | উদ্দেশ্য |
|
|
71
|
+
|------|-----------|-----------|----------|
|
|
72
|
+
| L1 সিস্টেম জ্ঞান | `knowledge/techs/` | টেক স্ট্যাক, আর্কিটেকচার, কনভেনশন | AI প্রকল্পের প্রযুক্তিগত সীমানা বোঝে |
|
|
73
|
+
| L2 ব্যবসায়িক জ্ঞান | `knowledge/bizs/` | মডিউল ফাংশন, ব্যবসায়িক প্রবাহ, সত্তা | AI ব্যবসায়িক যুক্তি বোঝে |
|
|
74
|
+
| L3 ইটারেশন আর্টিফ্যাক্ট | `iterations/iXXX/` | PRD, ডিজাইন ডকুমেন্ট, পরীক্ষার প্রতিবেদন | বর্তমান প্রয়োজনীয়তার জন্য সম্পূর্ণ ট্রেসেবিলিটি চেইন |
|
|
75
|
+
|
|
76
|
+
### 🔄 চার-পর্যায়ের জ্ঞান পাইপলাইন
|
|
77
|
+
**স্বয়ংক্রিয় জ্ঞান উৎপাদন আর্কিটেকচার**, সোর্স কোড থেকে ব্যবসায়িক/প্রযুক্তিগত ডকুমেন্টেশনের স্বয়ংক্রিয় উৎপাদন:
|
|
78
|
+
```
|
|
79
|
+
পর্যায় 1: সোর্স কোড স্ক্যান → মডিউল তালিকা তৈরি
|
|
80
|
+
পর্যায় 2: সমান্তরাল বিশ্লেষণ → বৈশিষ্ট্য বের করা (মাল্টি-ওয়ার্কার সমান্তরাল)
|
|
81
|
+
পর্যায় 3: সমান্তরাল সারাংশ → মডিউল ওভারভিউ সম্পূর্ণ করা (মাল্টি-ওয়ার্কার সমান্তরাল)
|
|
82
|
+
পর্যায় 4: সিস্টেম সংযোজন → সিস্টেম প্যানোরামা তৈরি
|
|
83
|
+
```
|
|
84
|
+
- **পূর্ণ সিঙ্কronization** এবং **ইনক্রিমেন্টাল সিঙ্কronization** সমর্থন করে (Git diff ভিত্তিক)
|
|
85
|
+
- একজন ব্যক্তি অপ্টিমাইজ করে, টিম শেয়ার করে
|
|
86
|
+
|
|
87
|
+
### 🔧 Harness বাস্তবায়ন কাঠামো
|
|
88
|
+
**স্ট্যান্ডার্ডাইজড এক্সিকিউশন ফ্রেমওয়ার্ক**, ডিজাইন ডকুমেন্টকে নির্ভুলভাবে এক্সিকিউটেবল ডেভেলপমেন্ট নির্দেশাবলীতে রূপান্তর নিশ্চিত করে:
|
|
89
|
+
- **অপারেশন ম্যানুয়াল নীতি**: Skill হল SOP, ধাপগুলি স্পষ্ট, ধারাবাহিক, স্বয়ংসম্পূর্ণ
|
|
90
|
+
- **ইনপুট-আউটপুট চুক্তি**: ইন্টারফেস স্পষ্টভাবে সংজ্ঞায়িত, সিউডোকোডের মতো কঠোরভাবে কার্যকর
|
|
91
|
+
- **গ্র্যাডুয়াল ডিসক্লোজার আর্কিটেকচার**: তথ্য স্তরে স্তরে লোড হয়, একবারে অত্যধিক প্রসঙ্গ এড়ায়
|
|
92
|
+
- **সাব-এজেন্ট ডেলিগেশন**: জটিল কাজ স্বয়ংক্রিয়ভাবে বিভক্ত হয়, সমান্তরাল কার্যকরীকরণ গুণমান নিশ্চিত করে
|
|
93
|
+
|
|
94
|
+
---
|
|
95
|
+
|
|
44
96
|
## ৮টি মূল সমস্যা সমাধান
|
|
45
97
|
|
|
46
98
|
### ১. AI বিদ্যমান প্রকল্প ডকুমেন্টেশন উপেক্ষা করে (জ্ঞান ব্যবধান)
|
package/README.bs.md
CHANGED
|
@@ -41,6 +41,58 @@ Integracijom Agenata i Vještina u postojeće projekte, timovi mogu brzo inicija
|
|
|
41
41
|
|
|
42
42
|
---
|
|
43
43
|
|
|
44
|
+
## ✨ Ključne Karakteristike
|
|
45
|
+
|
|
46
|
+
### 🏭 Virtuelni Softverski Tim
|
|
47
|
+
Jednoklik generisanje **7 profesionalnih uloga Agenata** + **30+ tokova rada Vještina**, izgradnja kompletnog virtuelnog softverskog tima:
|
|
48
|
+
- **Team Leader** - Globalno planiranje i upravljanje iteracijama
|
|
49
|
+
- **Product Manager** - Analiza zahtjeva i PRD output
|
|
50
|
+
- **Feature Designer** - Dizajn funkcionalnosti + API ugovori
|
|
51
|
+
- **System Designer** - Dizajn sistema Frontend/Backend/Mobilni/Desktop
|
|
52
|
+
- **System Developer** - Multiplatformski paralelni razvoj
|
|
53
|
+
- **Test Manager** - Koordinacija testiranja u tri faze
|
|
54
|
+
- **Task Worker** - Paralelno izvršavanje podzadataka
|
|
55
|
+
|
|
56
|
+
### 📐 ISA-95 Šestostepeno Modeliranje
|
|
57
|
+
Bazirano na međunarodnoj metodologiji modeliranja **ISA-95**, standardizacija transformacije poslovnih zahtjeva u softverske sisteme:
|
|
58
|
+
```
|
|
59
|
+
Domain Descriptions → Functions in Domains → Functions of Interest
|
|
60
|
+
↓ ↓ ↓
|
|
61
|
+
Information Flows → Categories of Information → Information Descriptions
|
|
62
|
+
```
|
|
63
|
+
- Svaki stepen odgovara specifičnim UML dijagramima (use case, sequence, class diagrams)
|
|
64
|
+
- Poslovni zahtjevi se "pročišćavaju korak po korak", bez gubitka informacija
|
|
65
|
+
- Rezultati su direktno upotrebljivi za razvoj
|
|
66
|
+
|
|
67
|
+
### 📚 Sistem Baze Znanja
|
|
68
|
+
Troslojna arhitektura baze znanja koja osigurava da AI uvijek radi bazirano na "jednom izvoru istine":
|
|
69
|
+
|
|
70
|
+
| Sloj | Direktorij | Sadržaj | Svrha |
|
|
71
|
+
|------|------------|---------|-------|
|
|
72
|
+
| L1 Sistemsko Znanje | `knowledge/techs/` | Tech stack, arhitektura, konvencije | AI razumije tehničke granice projekta |
|
|
73
|
+
| L2 Poslovno Znanje | `knowledge/bizs/` | Funkcionalnosti modula, poslovni tokovi, entiteti | AI razumije poslovnu logiku |
|
|
74
|
+
| L3 Artefakti Iteracija | `iterations/iXXX/` | PRD, dizajn dokumenti, test izvještaji | Kompletni lanac praćenja za trenutne zahtjeve |
|
|
75
|
+
|
|
76
|
+
### 🔄 Četvorostepeni Pipeline Znanja
|
|
77
|
+
**Automatizovana arhitektura generisanja znanja**, automatsko generisanje poslovne/tehničke dokumentacije iz izvornog koda:
|
|
78
|
+
```
|
|
79
|
+
Stepen 1: Skeniranje izvornog koda → Generisanje liste modula
|
|
80
|
+
Stepen 2: Paralelna analiza → Ekstrakcija funkcionalnosti (multi-Worker paralelno)
|
|
81
|
+
Stepen 3: Paralelno sažimanje → Dopuna pregleda modula (multi-Worker paralelno)
|
|
82
|
+
Stepen 4: Sistemska agregacija → Generisanje panorame sistema
|
|
83
|
+
```
|
|
84
|
+
- Podržava **potpunu sinhronizaciju** i **inkrementalnu sinhronizaciju** (bazirano na Git diff)
|
|
85
|
+
- Jedna osoba optimizira, tim dijeli
|
|
86
|
+
|
|
87
|
+
### 🔧 Harness Okvir za Praktičnu Implementaciju
|
|
88
|
+
**Standardizovani okvir za izvršavanje**, osigurava preciznu transformaciju dizajn dokumenata u izvršne razvojne instrukcije:
|
|
89
|
+
- **Princip operativnog priručnika**: Skill je SOP, koraci jasni, kontinuirani, samodovoljni
|
|
90
|
+
- **Ugovor o ulazima i izlazima**: jasno definisani interfejsi, strogo izvršavanje poput pseudokoda
|
|
91
|
+
- **Arhitektura postepenog otkrivanja**: informacije se učitavaju u slojevima, izbjegava preopterećenje konteksta
|
|
92
|
+
- **Delegiranje sub-agenta**: kompleksni zadaci se automatski dijele, paralelno izvršavanje osigurava kvalitet
|
|
93
|
+
|
|
94
|
+
---
|
|
95
|
+
|
|
44
96
|
## Rješavanje 8 ključnih problema
|
|
45
97
|
|
|
46
98
|
### 1. AI ignoriše postojeću dokumentaciju projekta (jaz u znanju)
|
|
@@ -189,7 +241,7 @@ speccrew init --ide claude
|
|
|
189
241
|
|
|
190
242
|
Nakon inicijalizacije, u projektu će biti generisano:
|
|
191
243
|
- `.qoder/agents/` / `.cursor/agents/` / `.claude/agents/` — 7 definicija uloga Agenata
|
|
192
|
-
- `.qoder/skills/` / `.cursor/skills/` / `.claude/skills/` —
|
|
244
|
+
- `.qoder/skills/` / `.cursor/skills/` / `.claude/skills/` — 30+ tokova rada Vještina
|
|
193
245
|
- `speccrew-workspace/` — Radni prostor (direktoriji iteracija, baza znanja, predlošci dokumenata)
|
|
194
246
|
- `.speccrewrc` — Konfiguracioni fajl SpecCrew
|
|
195
247
|
|
|
@@ -213,7 +265,22 @@ Prati standardni inženjerski tok rada korak po korak:
|
|
|
213
265
|
|
|
214
266
|
> Rezultati svake faze zahtijevaju ljudsku potvrdu prije prelaska na sljedeću fazu.
|
|
215
267
|
|
|
216
|
-
### 4.
|
|
268
|
+
### 4. Ažuriranje SpecCrew-a
|
|
269
|
+
|
|
270
|
+
Kada SpecCrew objavi novu verziju, potrebne su dvije koraka za dovršetak ažuriranja:
|
|
271
|
+
|
|
272
|
+
```bash
|
|
273
|
+
# Step 1: 更新全局 CLI 工具到最新版本
|
|
274
|
+
npm install -g speccrew@latest
|
|
275
|
+
|
|
276
|
+
# Step 2: 同步项目中的 Agents 和 Skills 到最新版本
|
|
277
|
+
cd /path/to/your-project
|
|
278
|
+
speccrew update
|
|
279
|
+
```
|
|
280
|
+
|
|
281
|
+
> **Napomena**: `npm install -g speccrew@latest` ažurira sam CLI alat, dok `speccrew update` ažurira datoteke definicija Agenta i Skillova u projektu. Oba koraka su potrebna za potpuno ažuriranje.
|
|
282
|
+
|
|
283
|
+
### 5. Druge CLI komande
|
|
217
284
|
|
|
218
285
|
```bash
|
|
219
286
|
speccrew list # Izlistaj instalirane agente i vještine
|
|
@@ -239,7 +306,7 @@ your-project/
|
|
|
239
306
|
│ │ ├── speccrew-system-developer.md # System Developer: Paralelni razvoj po platformama
|
|
240
307
|
│ │ ├── speccrew-test-manager.md # Test Manager: Koordinacija trofaznog testiranja
|
|
241
308
|
│ │ └── speccrew-task-worker.md # Task Worker: Paralelno izvršavanje podzadataka
|
|
242
|
-
│ └── skills/ #
|
|
309
|
+
│ └── skills/ # 30+ Vještina (grupisanih po funkciji)
|
|
243
310
|
│ ├── speccrew-pm-*/ # Upravljanje proizvodom (analiza zahtjeva, evaluacija)
|
|
244
311
|
│ ├── speccrew-fd-*/ # Feature Design (Feature Design, API ugovor)
|
|
245
312
|
│ ├── speccrew-sd-*/ # System Design (frontend/backend/mobile/desktop)
|
package/README.da.md
CHANGED
|
@@ -41,6 +41,58 @@ Ved at integrere Agents og Skills i eksisterende projekter kan teams hurtigt ini
|
|
|
41
41
|
|
|
42
42
|
---
|
|
43
43
|
|
|
44
|
+
## ✨ Nøglefunktioner
|
|
45
|
+
|
|
46
|
+
### 🏭 Virtuelt Softwareteam
|
|
47
|
+
Et-kliks generering af **7 professionelle Agent-roller** + **30+ Skill-workflows**, opbygning af et komplet virtuelt softwareteam:
|
|
48
|
+
- **Team Leader** - Global planlægning og iterationshåndtering
|
|
49
|
+
- **Product Manager** - Kravanalyse og PRD-output
|
|
50
|
+
- **Feature Designer** - Feature-design + API-kontrakter
|
|
51
|
+
- **System Designer** - Frontend/Backend/Mobil/Desktop-systemdesign
|
|
52
|
+
- **System Developer** - Multiplatform-paralleludvikling
|
|
53
|
+
- **Test Manager** - Trefaset-testkoordinering
|
|
54
|
+
- **Task Worker** - Parallel underopgaveudførelse
|
|
55
|
+
|
|
56
|
+
### 📐 ISA-95 Sekstrins Modellering
|
|
57
|
+
Baseret på international **ISA-95** modelleringsmetodik, standardisering af transformationen fra forretningskrav til softwaresystemer:
|
|
58
|
+
```
|
|
59
|
+
Domain Descriptions → Functions in Domains → Functions of Interest
|
|
60
|
+
↓ ↓ ↓
|
|
61
|
+
Information Flows → Categories of Information → Information Descriptions
|
|
62
|
+
```
|
|
63
|
+
- Hvert trin svarer til specifikke UML-diagrammer (use case, sekvens, klassediagrammer)
|
|
64
|
+
- Forretningskrav "forfines trin for trin" uden informationstab
|
|
65
|
+
- Outputs er direkte brugbare til udvikling
|
|
66
|
+
|
|
67
|
+
### 📚 Videnbase-system
|
|
68
|
+
Trenivå videnbase-arkitektur, der sikrer, at AI altid arbejder baseret på "den eneste sandhedskilde":
|
|
69
|
+
|
|
70
|
+
| Niveau | Bibliotek | Indhold | Formål |
|
|
71
|
+
|--------|-----------|---------|--------|
|
|
72
|
+
| L1 Systemviden | `knowledge/techs/` | Tech-stack, arkitektur, konventioner | AI forstår projektets tekniske grænser |
|
|
73
|
+
| L2 Forretningsviden | `knowledge/bizs/` | Modulfunktioner, forretningsflows, enheder | AI forstår forretningslogik |
|
|
74
|
+
| L3 Iterationsartefakter | `iterations/iXXX/` | PRD, designdokumenter, testrapporter | Komplet sporbarhedskæde for aktuelle krav |
|
|
75
|
+
|
|
76
|
+
### 🔄 Firetrins Videnspipeline
|
|
77
|
+
**Automatiseret vidensgenereringsarkitektur**, automatisk generering af forretnings/teknisk dokumentation fra kildekode:
|
|
78
|
+
```
|
|
79
|
+
Trin 1: Scan kildekode → Generer modulliste
|
|
80
|
+
Trin 2: Parallel analyse → Uddrag features (multi-Worker-parallel)
|
|
81
|
+
Trin 3: Parallel opsummering → Færdiggør moduloversigter (multi-Worker-parallel)
|
|
82
|
+
Trin 4: Systemaggregering → Generer systempanorama
|
|
83
|
+
```
|
|
84
|
+
- Understøtter **fuld synkronisering** og **inkrementel synkronisering** (baseret på Git diff)
|
|
85
|
+
- En person optimerer, team deler
|
|
86
|
+
|
|
87
|
+
### 🔧 Harness Praktisk Implementeringsramme
|
|
88
|
+
**Standardiseret eksekveringsramme**, sikrer præcis transformation af designdokumenter til eksekverbare udviklingsinstruktioner:
|
|
89
|
+
- **Operationsmanual-princip**: Skill er SOP, trin er klare, kontinuerlige, selvstændige
|
|
90
|
+
- **Input-output-kontrakt**: grænseflader tydeligt defineret, streng eksekvering som pseudokode
|
|
91
|
+
- **Gradvis offentliggørelsesarkitektur**: information indlæses lag for lag, undgår engangskontekst-overbelastning
|
|
92
|
+
- **Underagent-delegering**: komplekse opgaver opdeles automatisk, parallelt eksekvering sikrer kvalitet
|
|
93
|
+
|
|
94
|
+
---
|
|
95
|
+
|
|
44
96
|
## 8 Kerne-problemer Løst
|
|
45
97
|
|
|
46
98
|
### 1. AI Ignorerer Eksisterende Projektdokumentation (Videnskløft)
|
|
@@ -189,7 +241,7 @@ speccrew init --ide claude
|
|
|
189
241
|
|
|
190
242
|
Efter initialisering vil følgende blive genereret i dit projekt:
|
|
191
243
|
- `.qoder/agents/` / `.cursor/agents/` / `.claude/agents/` — 7 Agent-rolledefinitioner
|
|
192
|
-
- `.qoder/skills/` / `.cursor/skills/` / `.claude/skills/` —
|
|
244
|
+
- `.qoder/skills/` / `.cursor/skills/` / `.claude/skills/` — 30+ Skill-workflows
|
|
193
245
|
- `speccrew-workspace/` — Arbejdsområde (iterationsmapper, vidensbase, dokumentskabeloner)
|
|
194
246
|
- `.speccrewrc` — SpecCrew-konfigurationsfil
|
|
195
247
|
|
|
@@ -213,7 +265,22 @@ Følg standard engineering-workflow trin for trin:
|
|
|
213
265
|
|
|
214
266
|
> Hver phases leverancer kræver menneskelig bekræftelse før fortsættelse til næste fase.
|
|
215
267
|
|
|
216
|
-
### 4.
|
|
268
|
+
### 4. Opdatering af SpecCrew
|
|
269
|
+
|
|
270
|
+
Når SpecCrew udgiver en ny version, kræves der to trin for at fuldføre opdateringen:
|
|
271
|
+
|
|
272
|
+
```bash
|
|
273
|
+
# Step 1: 更新全局 CLI 工具到最新版本
|
|
274
|
+
npm install -g speccrew@latest
|
|
275
|
+
|
|
276
|
+
# Step 2: 同步项目中的 Agents 和 Skills 到最新版本
|
|
277
|
+
cd /path/to/your-project
|
|
278
|
+
speccrew update
|
|
279
|
+
```
|
|
280
|
+
|
|
281
|
+
> **Bemærkning**: `npm install -g speccrew@latest` opdaterer selve CLI-værktøjet, mens `speccrew update` opdaterer Agent- og Skill-definitionsfilerne i projektet. Begge trin skal udføres for at fuldføre den komplette opdatering.
|
|
282
|
+
|
|
283
|
+
### 5. Andre CLI-kommandoer
|
|
217
284
|
|
|
218
285
|
```bash
|
|
219
286
|
speccrew list # List installerede agents og skills
|
|
@@ -239,7 +306,7 @@ your-project/
|
|
|
239
306
|
│ │ ├── speccrew-system-developer.md # System Developer: Parallelt udvikling pr. platform
|
|
240
307
|
│ │ ├── speccrew-test-manager.md # Test Manager: Tre-fase testkoordinering
|
|
241
308
|
│ │ └── speccrew-task-worker.md # Opgavemedarbejder: Parallelt underopgave-eksekvering
|
|
242
|
-
│ └── skills/ #
|
|
309
|
+
│ └── skills/ # 30+ Skills (grupperet efter funktion)
|
|
243
310
|
│ ├── speccrew-pm-*/ # Produktstyring (kravanalyse, evaluering)
|
|
244
311
|
│ ├── speccrew-fd-*/ # Feature Design (Feature Design, API Kontrakt)
|
|
245
312
|
│ ├── speccrew-sd-*/ # System Design (frontend/backend/mobil/desktop)
|
package/README.de.md
CHANGED
|
@@ -41,6 +41,58 @@ Durch die Integration von Agents und Skills in bestehende Projekte können Teams
|
|
|
41
41
|
|
|
42
42
|
---
|
|
43
43
|
|
|
44
|
+
## ✨ Hauptmerkmale
|
|
45
|
+
|
|
46
|
+
### 🏭 Virtuelles Softwareteam
|
|
47
|
+
Ein-Klick-Generierung von **7 professionellen Agent-Rollen** + **30+ Skill-Workflows**, Aufbau eines vollständigen virtuellen Softwareteams:
|
|
48
|
+
- **Team Leader** - Globale Planung und Iterationsmanagement
|
|
49
|
+
- **Product Manager** - Anforderungsanalyse und PRD-Erstellung
|
|
50
|
+
- **Feature Designer** - Feature-Design + API-Verträge
|
|
51
|
+
- **System Designer** - Frontend/Backend/Mobile/Desktop-Systemdesign
|
|
52
|
+
- **System Developer** - Multiplattform-Parallelentwicklung
|
|
53
|
+
- **Test Manager** - Dreiphasen-Testkoordination
|
|
54
|
+
- **Task Worker** - Parallele Unteraufgabenausführung
|
|
55
|
+
|
|
56
|
+
### 📐 ISA-95 Sechs-Stufen-Modellierung
|
|
57
|
+
Basierend auf der internationalen **ISA-95** Modellierungsmethodik, Standardisierung der Transformation von Geschäftsanforderungen zu Softwaresystemen:
|
|
58
|
+
```
|
|
59
|
+
Domain Descriptions → Functions in Domains → Functions of Interest
|
|
60
|
+
↓ ↓ ↓
|
|
61
|
+
Information Flows → Categories of Information → Information Descriptions
|
|
62
|
+
```
|
|
63
|
+
- Jede Stufe entspricht bestimmten UML-Diagrammen (Anwendungsfälle, Sequenz, Klassendiagramme)
|
|
64
|
+
- Geschäftsanforderungen werden "schrittweise verfeinert", ohne Informationsverlust
|
|
65
|
+
- Ausgaben sind direkt für die Entwicklung verwendbar
|
|
66
|
+
|
|
67
|
+
### 📚 Wissensdatenbank-System
|
|
68
|
+
Dreischichtige Wissensdatenbank-Architektur, die sicherstellt, dass KI immer basierend auf der "einzigen Wahrheitsquelle" arbeitet:
|
|
69
|
+
|
|
70
|
+
| Ebene | Verzeichnis | Inhalt | Zweck |
|
|
71
|
+
|-------|-------------|---------|----------|
|
|
72
|
+
| L1 Systemwissen | `knowledge/techs/` | Tech-Stack, Architektur, Konventionen | KI versteht technische Projektgrenzen |
|
|
73
|
+
| L2 Geschäftswissen | `knowledge/bizs/` | Modulfeatures, Geschäftsabläufe, Entitäten | KI versteht Geschäftslogik |
|
|
74
|
+
| L3 Iterationsartefakte | `iterations/iXXX/` | PRD, Design-Dokumente, Testberichte | Vollständige Rückverfolgbarkeitskette für aktuelle Anforderungen |
|
|
75
|
+
|
|
76
|
+
### 🔄 Vier-Stufen-Wissenspipeline
|
|
77
|
+
**Automatisierte Wissensgenerierungsarchitektur**, automatische Generierung von Geschäfts-/Technikdokumentation aus Quellcode:
|
|
78
|
+
```
|
|
79
|
+
Stufe 1: Quellcode scannen → Modulliste generieren
|
|
80
|
+
Stufe 2: Parallele Analyse → Features extrahieren (Multi-Worker-parallel)
|
|
81
|
+
Stufe 3: Parallele Zusammenfassung → Modulübersichten vervollständigen (Multi-Worker-parallel)
|
|
82
|
+
Stufe 4: Systemaggregation → Systempanorama generieren
|
|
83
|
+
```
|
|
84
|
+
- Unterstützt **Voll-Synchronisierung** und **Inkrementelle Synchronisierung** (basierend auf Git diff)
|
|
85
|
+
- Eine Person optimiert, Team teilt
|
|
86
|
+
|
|
87
|
+
### 🔧 Harness Praktisches Implementierungsframework
|
|
88
|
+
**Standardisiertes Ausführungsframework**, stellt sicher, dass Design-Dokumente präzise in ausführbare Entwicklungsanweisungen umgewandelt werden:
|
|
89
|
+
- **Betriebsanleitungs-Prinzip**: Skill ist SOP, Schritte sind klar, fortlaufend, in sich geschlossen
|
|
90
|
+
- **Input-Output-Vertrag**: Schnittstellen klar definiert, strenge Ausführung wie Pseudocode
|
|
91
|
+
- **Progressive-Disclosure-Architektur**: Informationen werden schichtweise geladen, vermeidet einmalige Kontextüberlastung
|
|
92
|
+
- **Sub-Agent-Delegierung**: komplexe Aufgaben werden automatisch aufgeteilt, parallele Ausführung sichert Qualität
|
|
93
|
+
|
|
94
|
+
---
|
|
95
|
+
|
|
44
96
|
## 8 Kernprobleme gelöst
|
|
45
97
|
|
|
46
98
|
### 1. KI ignoriert bestehende Projektdokumentation (Wissenslücke)
|
|
@@ -189,7 +241,7 @@ speccrew init --ide claude
|
|
|
189
241
|
|
|
190
242
|
Nach der Initialisierung werden in Ihrem Projekt generiert:
|
|
191
243
|
- `.qoder/agents/` / `.cursor/agents/` / `.claude/agents/` — 7 Agent-Rollendefinitionen
|
|
192
|
-
- `.qoder/skills/` / `.cursor/skills/` / `.claude/skills/` —
|
|
244
|
+
- `.qoder/skills/` / `.cursor/skills/` / `.claude/skills/` — 30+ Skill-Workflows
|
|
193
245
|
- `speccrew-workspace/` — Workspace (Iterationsverzeichnisse, Wissensbasis, Dokumentvorlagen)
|
|
194
246
|
- `.speccrewrc` — SpecCrew-Konfigurationsdatei
|
|
195
247
|
|
|
@@ -213,7 +265,22 @@ Folgen Sie dem Standard-Engineering-Workflow Schritt für Schritt:
|
|
|
213
265
|
|
|
214
266
|
> Die Liefergegenstände jeder Phase erfordern menschliche Bestätigung vor dem Fortfahren zur nächsten Phase.
|
|
215
267
|
|
|
216
|
-
### 4.
|
|
268
|
+
### 4. SpecCrew Aktualisieren
|
|
269
|
+
|
|
270
|
+
Wenn SpecCrew eine neue Version veröffentlicht, sind zwei Schritte erforderlich, um das Update abzuschließen:
|
|
271
|
+
|
|
272
|
+
```bash
|
|
273
|
+
# Step 1: 更新全局 CLI 工具到最新版本
|
|
274
|
+
npm install -g speccrew@latest
|
|
275
|
+
|
|
276
|
+
# Step 2: 同步项目中的 Agents 和 Skills 到最新版本
|
|
277
|
+
cd /path/to/your-project
|
|
278
|
+
speccrew update
|
|
279
|
+
```
|
|
280
|
+
|
|
281
|
+
> **Hinweis**: `npm install -g speccrew@latest` aktualisiert das CLI-Tool selbst, während `speccrew update` die Agent- und Skill-Definitionsdateien im Projekt aktualisiert. Beide Schritte müssen ausgeführt werden, um das vollständige Update abzuschließen.
|
|
282
|
+
|
|
283
|
+
### 5. Andere CLI-Befehle
|
|
217
284
|
|
|
218
285
|
```bash
|
|
219
286
|
speccrew list # Installierte Agents und Skills auflisten
|
|
@@ -239,7 +306,7 @@ your-project/
|
|
|
239
306
|
│ │ ├── speccrew-system-developer.md # System Developer: Parallele Entwicklung nach Plattform
|
|
240
307
|
│ │ ├── speccrew-test-manager.md # Test Manager: Dreiphasige Testkoordination
|
|
241
308
|
│ │ └── speccrew-task-worker.md # Task Worker: Parallele Unteraufgabenausführung
|
|
242
|
-
│ └── skills/ #
|
|
309
|
+
│ └── skills/ # 30+ Skills (nach Funktion gruppiert)
|
|
243
310
|
│ ├── speccrew-pm-*/ # Produktmanagement (Anforderungsanalyse, Bewertung)
|
|
244
311
|
│ ├── speccrew-fd-*/ # Feature Design (Feature Design, API Vertrag)
|
|
245
312
|
│ ├── speccrew-sd-*/ # System Design (Frontend/Backend/Mobile/Desktop)
|
package/README.el.md
CHANGED
|
@@ -41,6 +41,58 @@
|
|
|
41
41
|
|
|
42
42
|
---
|
|
43
43
|
|
|
44
|
+
## ✨ Βασικά Χαρακτηριστικά
|
|
45
|
+
|
|
46
|
+
### 🏭 Εικονική Ομάδα Λογισμικού
|
|
47
|
+
Δημιουργία με ένα κλικ **7 επαγγελματικών ρόλων Agent** + **30+ ροών εργασίας Skills**, δημιουργία μιας πλήρους εικονικής ομάδας λογισμικού:
|
|
48
|
+
- **Team Leader** - Παγκόσμιος προγραμματισμός και διαχείριση επαναλήψεων
|
|
49
|
+
- **Product Manager** - Ανάλυση απαιτήσεων και έξοδος PRD
|
|
50
|
+
- **Feature Designer** - Σχεδιασμός λειτουργιών + συμβόλαια API
|
|
51
|
+
- **System Designer** - Σχεδιασμός συστημάτων Frontend/Backend/Mobile/Desktop
|
|
52
|
+
- **System Developer** - Πολλαπλή πλατφόρμα παράλληλης ανάπτυξης
|
|
53
|
+
- **Test Manager** - Συντονισμός δοκιμών τριών φάσεων
|
|
54
|
+
- **Task Worker** - Παράλληλη εκτέλεση δευτερευόντων εργασιών
|
|
55
|
+
|
|
56
|
+
### 📐 Μοντελοποίηση ISA-95 Έξι Σταδίων
|
|
57
|
+
Βασισμένο στη διεθνή μεθοδολογία μοντελοποίησης **ISA-95**, τυποποίηση του μετασχηματισμού επιχειρηματικών απαιτήσεων σε συστήματα λογισμικού:
|
|
58
|
+
```
|
|
59
|
+
Domain Descriptions → Functions in Domains → Functions of Interest
|
|
60
|
+
↓ ↓ ↓
|
|
61
|
+
Information Flows → Categories of Information → Information Descriptions
|
|
62
|
+
```
|
|
63
|
+
- Κάθε στάδιο αντιστοιχεί σε συγκεκριμένα διαγράμματα UML (use case, sequence, class)
|
|
64
|
+
- Οι επιχειρηματικές απαιτήσεις "εξευγενίζονται βήμα προς βήμα", χωρίς απώλεια πληροφοριών
|
|
65
|
+
- Τα αποτελέσματα είναι άμεσα χρησιμοποιήσιμα για ανάπτυξη
|
|
66
|
+
|
|
67
|
+
### 📚 Σύστημα Βάσης Γνώσης
|
|
68
|
+
Τριεπίπεδη αρχιτεκτονική βάσης γνώσης που διασφαλίζει ότι η AI εργάζεται πάντα με βάση την "μοναδική πηγή αλήθειας":
|
|
69
|
+
|
|
70
|
+
| Επίπεδο | Κατάλογος | Περιεχόμενο | Σκοπός |
|
|
71
|
+
|---------|-----------|-------------|--------|
|
|
72
|
+
| L1 Γνώση Συστήματος | `knowledge/techs/` | Tech stack, αρχιτεκτονική, συμβάσεις | Η AI καταλαβαίνει τα τεχνικά όρια του έργου |
|
|
73
|
+
| L2 Επιχειρηματική Γνώση | `knowledge/bizs/` | Λειτουργίες αρθρωμάτων, επιχειρηματικές ροές, οντότητες | Η AI καταλαβαίνει την επιχειρηματική λογική |
|
|
74
|
+
| L3 Τεχνουργήματα Επανάληψης | `iterations/iXXX/` | PRD, έγγραφα σχεδιασμού, αναφορές δοκιμών | Πλήρης αλυσίδα ιχνηλασιμότητας για τρέχουσες απαιτήσεις |
|
|
75
|
+
|
|
76
|
+
### 🔄 Αγωγός Γνώσης Τεσσάρων Σταδίων
|
|
77
|
+
**Αυτοματοποιημένη αρχιτεκτονική δημιουργίας γνώσης**, αυτόματη δημιουργία επιχειρηματικής/τεχνικής τεκμηρίωσης από πηγαίο κώδικα:
|
|
78
|
+
```
|
|
79
|
+
Στάδιο 1: Σάρωση πηγαίου κώδικα → Δημιουργία λίστας αρθρωμάτων
|
|
80
|
+
Στάδιο 2: Παράλληλη ανάλυση → Εξαγωγή λειτουργιών (πολλαπλοί Worker παράλληλα)
|
|
81
|
+
Στάδιο 3: Παράλληλη σύνοψη → Ολοκλήρωση επισκοπήσεων αρθρωμάτων (πολλαπλοί Worker παράλληλα)
|
|
82
|
+
Στάδιο 4: Συσσωμάτωση συστήματος → Δημιουργία πανοράματος συστήματος
|
|
83
|
+
```
|
|
84
|
+
- Υποστηρίζει **πλήρη συγχρονισμό** και **επαυξητικό συγχρονισμό** (βασισμένο σε Git diff)
|
|
85
|
+
- Ένα άτομο βελτιστοποιεί, η ομάδα μοιράζεται
|
|
86
|
+
|
|
87
|
+
### 🔧 Harness Πλαίσιο Υλοποίησης
|
|
88
|
+
**Τυποποιημένο πλαίσιο εκτέλεσης**, διασφαλίζει την ακριβή μετατροπή των εγγράφων σχεδιασμού σε εκτελέσιμες οδηγίες ανάπτυξης:
|
|
89
|
+
- **Αρχή εγχειριδίου λειτουργίας**: Το Skill είναι SOP, με σαφή, συνεχή και αυτοπεριεχόμενα βήματα
|
|
90
|
+
- **Συμβόλαιο εισόδου/εξόδου**: Καθορισμός σαφών διεπαφών, εκτέλεση με αυστηρότητα όπως ο ψευδοκώδικας
|
|
91
|
+
- **Αρχιτεκτονική σταδιακής αποκάλυψης**: Φόρτωση πληροφοριών σε επίπεδα, αποφυγή υπερφόρτωσης πλαισίου
|
|
92
|
+
- **Ανάθεση σε Sub-Agent**: Αυτόματος διαχωρισμός πολύπλοκων εργασιών, παράλληλη εκτέλεση για διασφάλιση ποιότητας
|
|
93
|
+
|
|
94
|
+
---
|
|
95
|
+
|
|
44
96
|
## 8 Βασικά Προβλήματα που Λύνονται
|
|
45
97
|
|
|
46
98
|
### 1. Η AI Αγνοεί την Υπάρχουσα Τεκμηρίωση Έργου (Κενό Γνώσης)
|
package/README.en.md
CHANGED
|
@@ -41,6 +41,58 @@ By integrating Agents and Skills into existing projects, teams can quickly initi
|
|
|
41
41
|
|
|
42
42
|
---
|
|
43
43
|
|
|
44
|
+
## ✨ Key Highlights
|
|
45
|
+
|
|
46
|
+
### 🏭 Virtual Software Team
|
|
47
|
+
One-click generation of **7 professional Agent roles** + **30+ Skill workflows**, building a complete virtual software team:
|
|
48
|
+
- **Team Leader** - Global scheduling and iteration management
|
|
49
|
+
- **Product Manager** - Requirements analysis and PRD output
|
|
50
|
+
- **Feature Designer** - Feature design + API contracts
|
|
51
|
+
- **System Designer** - Frontend/Backend/Mobile/Desktop system design
|
|
52
|
+
- **System Developer** - Multi-platform parallel development
|
|
53
|
+
- **Test Manager** - Three-phase test coordination
|
|
54
|
+
- **Task Worker** - Parallel sub-task execution
|
|
55
|
+
|
|
56
|
+
### 📐 ISA-95 Six-Stage Modeling
|
|
57
|
+
Based on international **ISA-95** modeling methodology, standardizing the transformation from business requirements to software systems:
|
|
58
|
+
```
|
|
59
|
+
Domain Descriptions → Functions in Domains → Functions of Interest
|
|
60
|
+
↓ ↓ ↓
|
|
61
|
+
Information Flows → Categories of Information → Information Descriptions
|
|
62
|
+
```
|
|
63
|
+
- Each stage corresponds to specific UML diagrams (use case, sequence, class diagrams)
|
|
64
|
+
- Business requirements are "refined step by step" with no information loss
|
|
65
|
+
- Outputs are directly usable for development
|
|
66
|
+
|
|
67
|
+
### 📚 Knowledge Base System
|
|
68
|
+
Three-tier knowledge base architecture ensuring AI always works based on the "single source of truth":
|
|
69
|
+
|
|
70
|
+
| Layer | Directory | Content | Purpose |
|
|
71
|
+
|-------|-----------|---------|----------|
|
|
72
|
+
| L1 System Knowledge | `knowledge/techs/` | Tech stack, architecture, conventions | AI understands project technical boundaries |
|
|
73
|
+
| L2 Business Knowledge | `knowledge/bizs/` | Module features, business flows, entities | AI understands business logic |
|
|
74
|
+
| L3 Iteration Artifacts | `iterations/iXXX/` | PRD, design docs, test reports | Complete traceability chain for current requirements |
|
|
75
|
+
|
|
76
|
+
### 🔄 Four-Stage Knowledge Pipeline
|
|
77
|
+
**Automated knowledge generation architecture**, auto-generating business/technical documentation from source code:
|
|
78
|
+
```
|
|
79
|
+
Stage 1: Scan source code → Generate module list
|
|
80
|
+
Stage 2: Parallel analysis → Extract features (multi-Worker parallel)
|
|
81
|
+
Stage 3: Parallel summarization → Complete module overviews (multi-Worker parallel)
|
|
82
|
+
Stage 4: System aggregation → Generate system panorama
|
|
83
|
+
```
|
|
84
|
+
- Supports **full sync** and **incremental sync** (based on Git diff)
|
|
85
|
+
- One person optimizes, team shares
|
|
86
|
+
|
|
87
|
+
### 🔧 Harness Execution Framework
|
|
88
|
+
**Standardized execution framework** ensuring design documents are accurately transformed into executable development instructions:
|
|
89
|
+
- **SOP Principle**: Skills as standard operating procedures—clear, continuous, self-contained steps
|
|
90
|
+
- **Input/Output Contracts**: Well-defined interfaces for rigorous, pseudocode-like execution
|
|
91
|
+
- **Progressive Disclosure**: Layered information architecture preventing context overload
|
|
92
|
+
- **Sub-Agent Dispatch**: Automatic task decomposition with parallel execution for quality assurance
|
|
93
|
+
|
|
94
|
+
---
|
|
95
|
+
|
|
44
96
|
## 8 Core Problems Solved
|
|
45
97
|
|
|
46
98
|
### 1. AI Ignores Existing Project Documentation (Knowledge Gap)
|
|
@@ -189,7 +241,7 @@ speccrew init --ide claude
|
|
|
189
241
|
|
|
190
242
|
After initialization, the following will be generated in your project:
|
|
191
243
|
- `.qoder/agents/` / `.cursor/agents/` / `.claude/agents/` — 7 Agent role definitions
|
|
192
|
-
- `.qoder/skills/` / `.cursor/skills/` / `.claude/skills/` —
|
|
244
|
+
- `.qoder/skills/` / `.cursor/skills/` / `.claude/skills/` — 30+ Skill workflows
|
|
193
245
|
- `speccrew-workspace/` — Workspace (iteration directories, knowledge base, document templates)
|
|
194
246
|
- `.speccrewrc` — SpecCrew configuration file
|
|
195
247
|
|
|
@@ -213,7 +265,22 @@ Follow the standard engineering workflow step by step:
|
|
|
213
265
|
|
|
214
266
|
> Each phase's deliverables require human confirmation before proceeding to the next phase.
|
|
215
267
|
|
|
216
|
-
### 4.
|
|
268
|
+
### 4. Update SpecCrew
|
|
269
|
+
|
|
270
|
+
When a new version of SpecCrew is released, complete the update in two steps:
|
|
271
|
+
|
|
272
|
+
```bash
|
|
273
|
+
# Step 1: Update the global CLI tool to the latest version
|
|
274
|
+
npm install -g speccrew@latest
|
|
275
|
+
|
|
276
|
+
# Step 2: Sync Agents and Skills in your project to the latest version
|
|
277
|
+
cd /path/to/your-project
|
|
278
|
+
speccrew update
|
|
279
|
+
```
|
|
280
|
+
|
|
281
|
+
> **Note**: `npm install -g speccrew@latest` updates the CLI tool itself, while `speccrew update` updates the Agent and Skill definition files in your project. Both steps are required for a complete update.
|
|
282
|
+
|
|
283
|
+
### 5. Other CLI Commands
|
|
217
284
|
|
|
218
285
|
```bash
|
|
219
286
|
speccrew list # List installed agents and skills
|
package/README.es.md
CHANGED
|
@@ -41,6 +41,58 @@ Al integrar Agentes y Skills en proyectos existentes, los equipos pueden inicial
|
|
|
41
41
|
|
|
42
42
|
---
|
|
43
43
|
|
|
44
|
+
## ✨ Características Principales
|
|
45
|
+
|
|
46
|
+
### 🏭 Equipo de Software Virtual
|
|
47
|
+
Generación con un clic de **7 roles de Agentes profesionales** + **30+ flujos de trabajo de Skills**, construyendo un equipo de software virtual completo:
|
|
48
|
+
- **Team Leader** - Planificación global y gestión de iteraciones
|
|
49
|
+
- **Product Manager** - Análisis de requisitos y generación de PRD
|
|
50
|
+
- **Feature Designer** - Diseño de funciones + contratos API
|
|
51
|
+
- **System Designer** - Diseño de sistemas Frontend/Backend/Móvil/Escritorio
|
|
52
|
+
- **System Developer** - Desarrollo paralelo multiplataforma
|
|
53
|
+
- **Test Manager** - Coordinación de pruebas en tres fases
|
|
54
|
+
- **Task Worker** - Ejecución paralela de subtareas
|
|
55
|
+
|
|
56
|
+
### 📐 Modelado ISA-95 de Seis Etapas
|
|
57
|
+
Basado en la metodología de modelado **ISA-95** estándar internacional, estandarizando la transformación de requisitos de negocio a sistemas de software:
|
|
58
|
+
```
|
|
59
|
+
Domain Descriptions → Functions in Domains → Functions of Interest
|
|
60
|
+
↓ ↓ ↓
|
|
61
|
+
Information Flows → Categories of Information → Information Descriptions
|
|
62
|
+
```
|
|
63
|
+
- Cada etapa corresponde a diagramas UML específicos (casos de uso, secuencia, clases)
|
|
64
|
+
- Los requisitos de negocio se "refinan paso a paso", sin pérdida de información
|
|
65
|
+
- Los resultados son directamente utilizables para el desarrollo
|
|
66
|
+
|
|
67
|
+
### 📚 Sistema de Base de Conocimientos
|
|
68
|
+
Arquitectura de base de conocimientos de tres niveles que asegura que la IA siempre trabaje basada en la "fuente única de verdad":
|
|
69
|
+
|
|
70
|
+
| Nivel | Directorio | Contenido | Propósito |
|
|
71
|
+
|-------|------------|-----------|-----------|
|
|
72
|
+
| L1 Conocimiento del Sistema | `knowledge/techs/` | Stack tecnológico, arquitectura, convenciones | IA entiende límites técnicos del proyecto |
|
|
73
|
+
| L2 Conocimiento de Negocio | `knowledge/bizs/` | Funciones de módulos, flujos de negocio, entidades | IA entiende lógica de negocio |
|
|
74
|
+
| L3 Artefactos de Iteración | `iterations/iXXX/` | PRD, documentos de diseño, informes de prueba | Cadena completa de trazabilidad para requisitos actuales |
|
|
75
|
+
|
|
76
|
+
### 🔄 Pipeline de Conocimientos de Cuatro Etapas
|
|
77
|
+
**Arquitectura de generación automatizada de conocimientos**, generando automáticamente documentación de negocio/técnica desde código fuente:
|
|
78
|
+
```
|
|
79
|
+
Etapa 1: Escanear código fuente → Generar lista de módulos
|
|
80
|
+
Etapa 2: Análisis paralelo → Extraer funciones (multi-Worker paralelo)
|
|
81
|
+
Etapa 3: Resumen paralelo → Completar vistas de módulos (multi-Worker paralelo)
|
|
82
|
+
Etapa 4: Agregación del sistema → Generar panorama del sistema
|
|
83
|
+
```
|
|
84
|
+
- Soporta **sincronización completa** e **sincronización incremental** (basado en Git diff)
|
|
85
|
+
- Una persona optimiza, el equipo comparte
|
|
86
|
+
|
|
87
|
+
### 🔧 Harness Marco de Implementación Práctica
|
|
88
|
+
**Marco de ejecución estandarizado**, asegura que los documentos de diseño se transformen con precisión en instrucciones de desarrollo ejecutables:
|
|
89
|
+
- **Principio del manual de operaciones**: Skill como SOP, pasos claros, continuos y autocontenidos
|
|
90
|
+
- **Contrato de entrada/salida**: Definición clara de interfaces, ejecución rigurosa como pseudocódigo
|
|
91
|
+
- **Arquitectura de revelación progresiva**: Carga de información en capas, evitando sobrecarga de contexto
|
|
92
|
+
- **Delegación de Sub-Agent**: División automática de tareas complejas, ejecución paralela para asegurar calidad
|
|
93
|
+
|
|
94
|
+
---
|
|
95
|
+
|
|
44
96
|
## 8 Problemas Principales Resueltos
|
|
45
97
|
|
|
46
98
|
### 1. La IA Ignora la Documentación Existente del Proyecto (Brecha de Conocimiento)
|
|
@@ -189,7 +241,7 @@ speccrew init --ide claude
|
|
|
189
241
|
|
|
190
242
|
Después de la inicialización, se generarán en su proyecto:
|
|
191
243
|
- `.qoder/agents/` / `.cursor/agents/` / `.claude/agents/` — 7 definiciones de roles Agent
|
|
192
|
-
- `.qoder/skills/` / `.cursor/skills/` / `.claude/skills/` —
|
|
244
|
+
- `.qoder/skills/` / `.cursor/skills/` / `.claude/skills/` — 30+ flujos de trabajo Skill
|
|
193
245
|
- `speccrew-workspace/` — Espacio de trabajo (directorios de iteración, base de conocimientos, plantillas de documentos)
|
|
194
246
|
- `.speccrewrc` — Archivo de configuración de SpecCrew
|
|
195
247
|
|
|
@@ -213,7 +265,22 @@ Siga el flujo de trabajo de ingeniería estándar paso a paso:
|
|
|
213
265
|
|
|
214
266
|
> Los entregables de cada fase requieren confirmación humana antes de proceder a la siguiente fase.
|
|
215
267
|
|
|
216
|
-
### 4.
|
|
268
|
+
### 4. Actualizar SpecCrew
|
|
269
|
+
|
|
270
|
+
Cuando SpecCrew lanza una nueva versión, se requieren dos pasos para completar la actualización:
|
|
271
|
+
|
|
272
|
+
```bash
|
|
273
|
+
# Step 1: 更新全局 CLI 工具到最新版本
|
|
274
|
+
npm install -g speccrew@latest
|
|
275
|
+
|
|
276
|
+
# Step 2: 同步项目中的 Agents 和 Skills 到最新版本
|
|
277
|
+
cd /path/to/your-project
|
|
278
|
+
speccrew update
|
|
279
|
+
```
|
|
280
|
+
|
|
281
|
+
> **Nota**: `npm install -g speccrew@latest` actualiza la herramienta CLI en sí, mientras que `speccrew update` actualiza los archivos de definición de Agent y Skill en el proyecto. Ambos pasos deben ejecutarse para completar la actualización completa.
|
|
282
|
+
|
|
283
|
+
### 5. Otros Comandos CLI
|
|
217
284
|
|
|
218
285
|
```bash
|
|
219
286
|
speccrew list # Listar agents y skills instalados
|
|
@@ -239,7 +306,7 @@ your-project/
|
|
|
239
306
|
│ │ ├── speccrew-system-developer.md # System Developer: Desarrollo paralelo por plataforma
|
|
240
307
|
│ │ ├── speccrew-test-manager.md # Test Manager: Coordinación de pruebas de tres fases
|
|
241
308
|
│ │ └── speccrew-task-worker.md # Task Worker: Ejecución paralela de subtareas
|
|
242
|
-
│ └── skills/ #
|
|
309
|
+
│ └── skills/ # 30+ Skills (agrupados por función)
|
|
243
310
|
│ ├── speccrew-pm-*/ # Gestión de Producto (análisis de requisitos, evaluación)
|
|
244
311
|
│ ├── speccrew-fd-*/ # Feature Design (Feature Design, Contrato API)
|
|
245
312
|
│ ├── speccrew-sd-*/ # System Design (frontend/backend/móvil/escritorio)
|