deckent — Product and Runtime Reference

Identity

Canonical ürün kimliği ve execution-authority tanımı: .deckent/workspace/IDENTITY.md. @file import çözmeyen host'lar bu dosyayı doğrudan açıp okur.

Aktif yön kararları taşındı — bu iç-pivot notu artık .analysis/hermes-vs-deckent-direction-decisions.md içinde tutuluyor (CLAUDE.md operating_rules'ta da özetlenir); DECKENT.md yalnız mekanik/işlevsel referans kalır, tekrar-eden iç-strateji notu barındırmaz.

Rules

Canonical authority chain Goal → Mission → Flow → Run → WorkItem → Attempt → Operation'dır; Brain bu authority'yi supervise eder, worker kendi scope veya authority'sini genişletemez
Workers stay within assigned scope (directories + filesWrite)
Auditor never writes source code
Her run typed terminal veya resumable settlement'a ulaşır; PAUSED/HOLD geçerli durumlar, belirsiz/stale running değildir
DIRECTIVES.md is the binding execution contract for an active run after owner/system instructions
Runtime limits and retention budgets come from effective config; this document does not freeze live values

Providers

Deckent's product contract is provider-neutral. Provider identities and adapters are discovered from effective config + the runtime registry; this file is not a provider catalog
Brain/worker/auditor provider, exact model, effort and backend are resolved per role from effective config, registry and capability evidence
Role-aware fallback: provider_fallback.brain|worker|auditor → provider_fallback.global → legacy single fallback_provider; provider_fallback.auditor_provider selects the Auditor primary
Configured order is authoritative, but fallback is never an availability claim: auth, backend/model reachability, limit evidence and execution-budget admission must still pass
Model Registry: live/cached catalog plus bundled offline fallback — src/core/model-registry.ts; inspect with deckent models list
Runtime model identity is the exact registered provider API ID (id === apiId); aliases are migration inputs only and are rejected in authored tasks
Provider-agnostic selection uses model_strategy.brain_tier / model_strategy.worker_tier; tier resolution returns a registered exact API ID
Planning mode: brain_planning = 'ai' | 'structured' | 'auto'

Agents & Skills

Canlı agent listesi docs/reference/agents.md ve deckent agents; canlı skill listesi registry/deckent skills yüzeyidir. Bu belgede sayı tutulmaz
Agent pool: .deckent/agents/*/agent.json; retention/eviction effective config ve policy'den çözülür
Skill registry: .deckent/skills/*/skill.json — AST sandbox validation
Task routing: task-router.ts assigns agent + skills + provider per task

MCP Integration

Tool/resource sayıları bu belgede tutulmaz. Canonical canlı liste koddan auto-generated docs/reference/mcp-tools.md içindedir (npm run docs:ref)
MCP server entrypoint provider/host bağımsızdır: npx deckent-mcp; registration şekli seçili host adapterının capability/config yüzeyinden gelir

Memory V2 — DB-First Architecture

Storage: SQLite (better-sqlite3) — single source of truth, .md files are generated exports
Search: FTS5 full-text search with dual-layer Turkish normalize (TR/EN/DE %100 recall)
Product/user memory authority: .brain/memory.db (gitignored; asla routine cleanup hedefi değildir)
Deckent-dev dogfood core-memory authority: .deckent/docs/core-memory/MEMORY.md + aynı dizindeki referanslar; product memory değildir
Exports: .brain/exports/summary.md, decisions.md, memory.md, debt.md (git-tracked)
Schema: 5 tables (entries, tags, relations, entry_history, schema_version) + FTS5 virtual table
Brain auto-query: Task DNA → ilgili ADR/pattern/memory otomatik sorgulanır (PLAN, SPAWN, EVALUATE)
CLI: deckent recall "sorgu", deckent remember "not", deckent memory rebuild|export|stats
MCP: deckent_memory_query tool — cross-source hafıza arama
Config: .deckent/config.json → memory.backend, memory.search, memory.decay_after_sprints

Mandatory Architecture Rules

Generated canlı durum projection'ı: .brain/exports/summary.md (gerektiğinde doğrudan oku; runtime authority değildir).

Architecture Decision Records

.brain/memory.db = accepted ADR authority; docs/adr/*.md ve .brain/exports/decisions.md review/search projectionlarıdır
.deckent/decisions/*.json = SDL (Sprint Decision Log) — tactical decisions, audit trail, optional
Dependency execution davranışı effective config'ten okunur; dependency-aware scheduling ADR-G-026'ya, manuel dogfood recovery yalnız ADR-D-007'ye tabidir. Bu belge canlı boolean/default veya worker sayısı stamp etmez

Context

DIRECTIVES.md — yalnız DIRECTIVES-backed aktif run için bağlayıcı execution contract
.brain/exports/summary.md — generated status/knowledge projection
docs/reference/api-surface.md — transport ve runtime contract haritası

Environment

Boot

Bootstrap rehberi: .deckent/workspace/BOOT.md (host import çözmüyorsa doğrudan aç-oku).

Workflow Guide — Is Akisi Rehberi

Deckent'in Goal → Mission → Flow → Run authority zinciri çeşitli adapter yüzeylerinden yürütülebilir. Aşağıdaki sprint akışı, DIRECTIVES-backed CLI/MCP adapterının compatibility workflow'udur; ürünün bütünü veya tek execution authority'si değildir:

Closure OS governance yüzeyi (Phase-4 foundation): MASTER'dan ayrı, append-only hash-chained sidecar decision-ledger (Level×Lane sınıflandırma + admission + priority-karar authority'si); reviewed-parent trust-anchor, immutable historical snapshot bundle ve transactional four-view projection ile. Buildless gate npm run lint:gates içinde (scripts/lint-closure-dispositions.mjs); projector scripts/closure-ledger/project.mjs. Owner-verified public genesis trust anchor main'dedir (PR #127, commit 88637d5d6; private signer key repo DIŞINDA owner custody). Phase-5 CANLI (2026-08-17): writer + owner sign ceremony sevk edildi; ilk authenticated batch (dba89c03…, 2 event, owner-signed ed25519 receipt) append edildi, dört-view projection üretildi ve 8101+7140 MASTER settlement'ı bu batch'e bağlı receipt'lerle kapandı. Tam referans: docs/governance/closure-os-sidecar-ledger.md.

deckent_init — Projeyi başlat. .deckent/, .brain/, .tasks/ dizinlerini ve seçili host adapterının provider-neutral kural/reference projection'larını oluşturur. Desteklenen her environment/project scope'unda bir kez çalıştırılır.
deckent_set_directives — Sprint hedeflerini yaz. DIRECTIVES.md dosyasini gunceller. Asagidaki DIRECTIVES Format Rehberini kullanarak task'lari tanimla.
deckent_plan — Sprint planla. DIRECTIVES.md'yi okur, task JSON dosyalarini .tasks/ altinda olusturur. mode: 'ai' ile yapay zeka tabanli planlama, mode: 'structured' ile kural tabanli planlama.
deckent_start — Sprinti baslat. Worker'lari tmux veya subprocess olarak spawn eder, Auditor scan dongusu baslar. --dry-run ile task listesini goruntule, spawn etme.
deckent_status — Ilerlemeyi izle. Aktif worker'lar, tamamlanan task'lar, alertler ve coverage bilgisini gosterir. --watch ile canli izleme, --json ile ham JSON ciktisi.
deckent_review — Sprint sonucunu degerlendir. GO / NO_GO / GO_WITH_TECH_DEBT karari verir. Hangi task'larin basarisiz oldugunu ve tech debt birakilip birakilmayacagini gosterir.
deckent_retro — Retrospektif oku. RETRO.md ozetini, ogrenimleri ve bir sonraki sprint icin onerileri gosterir.
deckent_cleanup — Sprinti temizle. Task dosyalarini arsivler, kilitleri serbest birakir, tmux sessionlarini kapatir.

Native Terminal Work-Launching Flow — Terminal İş-Başlatma Akışı

Canonical Flow (terminal.run_flow_v2 ON)

When terminal.run_flow_v2 is enabled, the canonical entry point for launching work via the native terminal is deckent_propose_run. The flow is:

Natural Language Intent
        ↓
  deckent_propose_run
        ↓
   Brain Plan Preview
   (tasks, gates, digest)
        ↓
  In-Terminal Card
  (approve/reject UI)
        ↓
   Snapshot Start
   (worker spawn)
        ↓
 Correlated Result

Step-by-step:

Natural Language Intent — You describe the work in plain language: "Add TypeScript strict mode to src/core", "Fix flaky test in utils", etc.
deckent_propose_run Call — The terminal agent proposes a run with your intentSummary. This call:
- Generates a real Brain plan preview (task summaries, gate/policy results, content-addressed digest)
- Never starts anything and never requires a confirmation prompt
- Returns immediately with the preview compiled
Plan Preview Card — The terminal renders an in-terminal UI card showing:
- Proposed tasks and their summaries
- Any gate/policy checks (security, dependency, etc.)
- Content digest (hash of the plan)
- Your choices: Approve, Reject, or Edit
Human Approval — You review and approve the plan via the card (not a text prompt). This is the actual gate.
Snapshot Start — On approval, deckent snapshots the plan state and spawns workers.
Correlated Result — Workers report results linked back to the original proposal for audit and tracing.

Legacy Flow (terminal.run_flow_v2 OFF, or expert escape hatch)

The prior work-launching flow remains available as an expert low-level escape hatch:

DIRECTIVES.md
      ↓
 deckent_set_directives
      ↓
  DIRECTIVES written
      ↓
  deckent_plan
      ↓
  .tasks/ JSON created
      ↓
  deckent_start
      ↓
  Workers spawn

This flow is useful when:

You need precise manual control over task definitions and scope
You're debugging or iterating on task structure
terminal.run_flow_v2 is disabled (the default)

Explicit note: When terminal.run_flow_v2 is ON, this path is labeled an expert escape hatch in the native terminal tool descriptions, and deckent_propose_run is the canonical entry point. When the flag is OFF, the legacy flow is the only path available.

Configuration

{
  "terminal": {
    "run_flow_v2": false
  }
}

Set in .deckent/config.json (see deckent config set terminal.run_flow_v2 true via the CLI).

Default: false (legacy flow active)
Canonical mode: true (enables deckent_propose_run as the main entry point)

Traceability & Audit

Both flows produce a result record linked to the initiating act:

Canonical flow: Proposal ID → Plan preview digest → Approval event → Snapshot → Worker results
Legacy flow: DIRECTIVES version → Plan JSON → Start event → Worker results

The Brain's memory system (memory.db) records both paths equally for audit and learning.

DIRECTIVES Format Guide — DIRECTIVES Format Rehberi

DIRECTIVES.md dosyasi sprint hedeflerini tanimlar. Her task asagidaki formati izlemelidir:

# DIRECTIVES — Sprint NNN: Sprint Basligi

## Goal: Sprint amacini bir paragrafta acikla.

---

## Task 1: Task Basligi
- Model: <exact-registered-model-api-id>
- Provider: <registered-provider-id>
- Effort: normal
- Skills: typescript-expert
- Files: src/core/config.ts, src/core/types.ts
- Scope: src/core/

### Description
Task'in ne yapacagini detayli acikla. Hangi degisiklikler yapilacak,
hangi fonksiyonlar eklenecek/degistirilecek, neden gerekli oldugunu belirt.

**Kanit:** `grep "yeniOzellik" src/core/config.ts` → eklendi

**Test:** 3+ test (temel davranis, edge case, hata durumu)

---

## Task 2: Diger Task
- Model: <exact-registered-model-api-id>
- Provider: <registered-provider-id>
- Effort: low
- Skills: documentation-writer
- Files: README.md, docs/guide.md
- Scope: docs/

### Description
...

Alan Aciklamalari

Alan	Gecerli Degerler	Aciklama
Model	Kayıtlı tam provider API ID	Kullanılacak exact model; legacy alias'lar reddedilir, canlı katalog için `deckent models list`
Provider	Kayıtlı provider ID	Bu task'ı hangi provider adapterı koşsun (per-task override; erişilebilirlik kanıtı değildir)
Effort	low, normal, high	Tahmini is YUKU (timeout/butce/token-tahmin). Reasoning-derinligi DEGIL — onun icin ModelEffort
ModelEffort	Adapter capability'sinin kabul ettiği değer	Modelin reasoning DERİNLİĞİ; supported değer registry/adapter metadata'sından gelir. Effort (iş-yükü) ile karıştırma
Backend	Kayıtlı execution-backend ID	Task'ı belirli backend'e zorlar; provider/backend compatibility ve capability admission ayrıca doğrulanır
Auth	Kayıtlı auth-mode ID	Per-task auth modu; gerekli credential/entitlement adapter contractından çözülür
Skills	skill-id listesi	Uzmanlik alani (virgul ile ayir)
Files	dosya yollari	Degistirilecek dosyalar
Scope	dizin yollari	Izin verilen dizinler
Kanit	shell komutu	Tamamlanma kaniti
Test	test sayisi + aciklama	Beklenen testler

Sprint Lifecycle — Sprint Yasam Dongusu (run, eskiden "sprint")

Bir sprint 8 fazdan olusur:

PLAN → SPAWN → EXECUTE → EVALUATE → FIX → RETRO → DECAY → COMPLETE  <!-- cleanup = komut, faz değil (SprintPhase enum; hizalama 2026-08-01) -->

Faz	Aciklama	Sorumlu
PLAN	Brain DIRECTIVES'i okur, task JSON'larini olusturur	Brain
SPAWN	Worker'lar tmux/subprocess ile baslatilir, Auditor baslar	Brain
EXECUTE	Worker'lar task'lari uygular, heartbeat dosyalari yazar	Workers
EVALUATE	Brain sonuclari degerlendirir: GO / NO_GO / GO_WITH_TECH_DEBT	Brain
FIX	Basarisiz task'lar yeniden denenir (yapilandirilabilir timeout)	Brain + Workers
RETRO	Retrospektif yazilir: RETRO.md, sprint log guncellenir	Brain
DECAY	.brain/ bellek butcesi asildiysa eski satirlar temizlenir	Brain
CLEANUP	Task dosyalari arsivlenir, kilitler serbest, session'lar kapatilir	Brain

MCP Tool Reference — MCP Arac Referansi

Tam araç listesi (51 araç) koddan otomatik türetilir — bkz. docs/generated/tr/reference/mcp-tools.md (npm run docs:ref). Eskiden burada duran el-yazımı tablo drift etmişti (33 vs 35 vs 46); canonical generated referans tek-kaynaktır. Araç adları yukarıdaki ## MCP Integration bölümünde listelenir; somut parametre örnekleri aşağıdadır.

Parametre Ornekleri

// deckent_init
{ root: "/path/to/project", projectName: "my-app" }

// deckent_set_directives
{ content: "# DIRECTIVES — Sprint 001\n## Task 1: ...", root: "/path/to/project" }

// deckent_plan
{ mode: "ai", root: "/path/to/project" }
// mode: "ai" → registered provider/model ile LLM-assisted planlama
// mode: "structured" → Kural tabanli, deterministik
// mode: "auto" → Proje boyutuna gore otomatik sec

// deckent_start
{ sprintId: "sprint-001", dryRun: false, root: "/path/to/project" }

// deckent_status
{ watch: false, json: false, root: "/path/to/project" }

// deckent_config
{ action: "read", root: "/path/to/project" }
{ action: "set", key: "max_workers", value: "<positive-integer-or-auto>", root: "/path/to/project" }

// deckent_kill
{ target: "all", root: "/path/to/project" }
{ target: "worker", workerId: "w-001-003", root: "/path/to/project" }

approvals ve xverify kullanıcı akışı

approvals — runtime onay isteklerinin yönetimi:

deckent approvals list → bekleyen (pending) onay isteklerini listeler (read-only).
deckent approvals decide <requestId> --allow|--deny [--reason <text>] → kararı verir; bu mutation YALNIZ CLI'da, interactive live-auth arkasında yapılır.
MCP deckent_approvals yalnız read-only pending inbox'tur (canonical ApprovalBroker → list('pending')); MCP üzerinden allow/deny/decide/self-approval YOKTUR.

xverify — cross-provider host-adjudicated doğrulama:

deckent xverify … bir iddiayı ÜRETEN provider'dan FARKLI bir provider ile bağımsız doğrular (XVERIFY-PROVIDER-SEPARATION; same-provider self-verify yasaktır).
Host, CONFIRMED/REFUTED/UNCLEAR → ALLOW/NO-GO/HOLD disposition'ını türetir; assurance typed-host-adjudicated.
Kapanış (closure) YALNIZ şu zincirle olur: verdict + gerçek provider call + provider-reported usage + terminally-closed settlement + durable verdict receipt (cross-verify-verdict:sha256:…). HOLD/UNCLEAR kapanış DEĞİLDİR. Kanıtlanan Fable→Sol koşusu (CONFIRMED/ALLOW, total 60787 token) ve owner-bounded subscription budget için bkz. CLOSURE-OS-PRODUCT-TRANSITION-BRIEF.md §12.2.

MCP Resource Reference — MCP Kaynak Referansi

Resource	URI	Icerik	Aciklama
dashboard	`deckent://dashboard`	JSON	Aktif sprint durumu: worker'lar, fazlar, alertler, metrikler
directives	`deckent://directives`	Markdown	Mevcut DIRECTIVES.md icerigi
memory	`deckent://memory`	Markdown	Brain bellek ozeti: ogrenim, kararlar, desenler
debt	`deckent://debt`	Markdown	Teknik borc tablosu: acik ve cozulmus maddeler
config	`deckent://config`	JSON	Guncellenmis proje konfigurasyonu
retro	`deckent://retro`	Markdown	Son sprint retrospektif raporu
tasks	`deckent://tasks`	JSON	Mevcut sprint task listesi ve durumlari
agents	`deckent://agents`	JSON	Kayitli agent havuzu, istatistikler, kullanim oranlari

Parameter Reference — Parametre Referansi

Model

Model adları ve catalog üyeleri bu dosyada tutulmaz. Canonical kimlik id === apiId'dir. Güncel catalog presence için deckent models list; gerçek kullanılabilirlik için auth/account, backend/model reachability, entitlement, limit ve finite-budget evidence birlikte doğrulanır. Authored task exact registry ID kullanır; tier-based seçim exact ID'yi runtime'da çözer.

Tier

Tier	Aciklama
`premium_plus`	En yüksek capability sınıfı; exact model registry'den çözülür
`premium`	Karmaşık görev/mimari sınıfı; exact model registry'den çözülür
`standard`	Genel geliştirme sınıfı; exact model registry'den çözülür
`economy`	Düşük-maliyet/düşük-risk sınıfı; exact model registry'den çözülür

Effort

Deger	Aciklama	Ornek
`low`	<1 saat, minimal degisiklik	Yorum guncelleme, kucuk duzeltme
`normal`	1-3 saat, orta kapsamli	Yeni fonksiyon, test ekleme
`high`	3+ saat, buyuk degisiklik	Yeni modul, mimari degisiklik, refactor

Planning Mode

Deger	Aciklama
`ai`	AI tabanli planlama — DIRECTIVES'i yorumlar, akilli task bolunumu
`structured`	Kural tabanli — deterministik, hizli, AI API kullanmaz
`auto`	Proje boyutuna gore otomatik: kucuk→structured, buyuk→ai

Provider

Provider listesi registry/effective config'ten gelir. Her provider adapterı aynı contractı karşılar: exact identity, auth/account, reachability, entitlement/limit, usage/metering, execution backend, settlement ve gerektiğinde landing capability. Eksik capability sessiz fallback değil typed HOLD/UNSUPPORTED üretir.

Brain, Worker ve Auditor icin fallback sirasi ayri tanimlanabilir:

{
  "providers": {
    "brain": "<registered-provider-id>",
    "worker": "<registered-provider-id>"
  },
  "provider_fallback": {
    "brain": ["<registered-fallback-id>"],
    "worker": ["<registered-fallback-id>"],
    "auditor_provider": "<registered-provider-id>",
    "auditor": ["<registered-fallback-id>"],
    "global": ["<registered-fallback-id>"],
    "unattended": false
  }
}

Per-role zincir varsa global zincirin yerini alir; primary zincirden cikarilir ve duplicate'ler configured order korunarak tekillestirilir. Bu config yalniz aday sirasini tanimlar: bilinmeyen/stale limit veya kanitsiz reachability unattended fallback'i available yapmaz.

Error Resolution Guide — Hata Cozum Rehberi

Sprint Takildi / Dondu

# 1. Aktif worker'lari durdur
deckent kill --all

# 2. Task dosyalarini temizle
deckent cleanup

# 3. Codebase sagligini kontrol et
deckent doctor

# 4. Gerekirse yeniden baslat
deckent start

MCP Araciligi ile:

deckent_kill  → { target: "all" }
deckent_cleanup → { root: "." }
deckent_doctor → { root: "." }

Konfigurasyon Sorunu

# Mevcut konfigurasyon'u oku
deckent config read

# Belirli bir degeri guncelle
deckent config set max_workers <positive-integer-or-auto>
deckent config set providers.brain <registered-provider-id>
deckent config set routing_engine <registered-routing-engine-id>

Build Hatasi (tsc --noEmit)

Hata mesajini oku — hangi dosya, hangi satir
Ilgili tipleri kontrol et (src/core/*-types.ts)
Import yollarini dogrula (.js uzantisi gerekli — ESM)
tsc --noEmit tekrar calistir

Test Basarisizligi

Basarisiz test'in dosyasini tespit et
npx vitest run path/to/failing.test.ts ile izole calistir
Mock'larin guncellenmis export'lari icerip icermedigini kontrol et
npx vitest run — tam suite calistir

Worker .result Birakmadi (False NO_GO)

Worker tmux session'inda calisiyor olabilir: tmux ls ile kontrol et
.tasks/task-NNN.hb heartbeat dosyasini kontrol et — son timestamp
deckent status ile worker durumunu gozlemle
Gerekirse: deckent kill --worker w-NNN-NNN

MCP Server Eski Kod Calistiriyor

# Build et
tsc

# MCP server'i yeniden baslat (long-lived process cache'i temizler)
# Seçilen host adapterının documented MCP restart/reconnect akışını uygula

Built-in Agents (temsili alt-küme; canlı liste: `docs/reference/agents.md`)

ADR-G-023: Agent/skill taxonomy; test görevi task-bazlı yönetilir.

Agent	Uzmanlik	Aktivasyon
`security-auditor`	Guvenlik aciklari, OWASP top 10, auth	security/auth/vuln anahtar kelimeleri
`doc-writer`	README, JSDoc, API docs, CHANGELOG	docs/readme/comment anahtar kelimeleri
`bug-fixer`	Hata ayiklama, regression, hotfix	fix/bug/error/crash anahtar kelimeleri
`code-reviewer`	Kod kalitesi, best practices, PR review	review/quality/refactor anahtar kelimeleri
`refactorer`	Yeniden yapilandirma, temizlik, modernizasyon	refactor/cleanup/migrate anahtar kelimeleri
`api-builder`	REST API, OpenAPI, endpoint tasarimi	api/endpoint/route/schema anahtar kelimeleri
`performance-analyzer`	Profiling, optimizasyon, benchmark	perf/slow/optimize anahtar kelimeleri
`ci-guardian`	CI/CD saglik, test regresyon, build	ci/pipeline/test anahtar kelimeleri
`architect`	Sistem tasarimi, modul yonetimi, bagimlilk analizi	architecture/design/module anahtar kelimeleri
`architecture-planner`	Mimari planlama, ADR yazimi, yol haritasi	plan/roadmap/adr anahtar kelimeleri
`accessibility-auditor`	WCAG, a11y, erisilebilirlik denetimi	accessibility/a11y/wcag anahtar kelimeleri
`data-engineer`	Veri pipeline, ETL, veri modeli	data/pipeline/etl anahtar kelimeleri
`devops-engineer`	CI/CD, Docker, deployment, altyapi	devops/deploy/docker anahtar kelimeleri
`frontend-designer`	UI/UX, component tasarimi, responsive	frontend/ui/design anahtar kelimeleri
`migration-specialist`	Versiyon gecisi, framework migration	migration/upgrade/deprecation anahtar kelimeleri

Built-in Skills (temsili alt-küme; canlı liste: registry + `src/core/builtins/skills/`)

Skill	Aciklama
`typescript-expert`	TypeScript tip sistemi, ESM, generics, decorators
`testing-expert`	Vitest/Jest, mock'lama, coverage, test stratejisi
`documentation-writer`	Markdown, JSDoc, API docs, changelog
`security-specialist`	Guvenlik patternleri, input validasyon, kriptografi
`performance-optimizer`	Async optimizasyon, memory, profiling
`api-builder`	REST tasarimi, OpenAPI spec, versiyonlama
`devops-engineer`	GitHub Actions, Docker, deployment pipeline
`database-migration`	Query optimizasyon, migration, ORM
`react-specialist`	React, Vite, Tailwind, component mimari
`python-expert`	Python ekosistemi, FastAPI, veri islemleri
`ci-testing`	CI ortaminda test yurutme, regresyon algilama
`accessibility-expert`	WCAG standartlari, a11y test, erisilebilirlik
`code-simplifier`	Kod sadestirme, karmasiklik azaltma, temizlik
`docker-expert`	Dockerfile, compose, container optimizasyon
`frontend-design`	UI component, CSS, responsive tasarim
`git-expert`	Git is akisi, branch stratejisi, merge yonetimi
`graphql-expert`	GraphQL schema, resolver, subscription
`migration-expert`	Framework gecisi, versiyon yukseltme, API migration
`monorepo-expert`	Monorepo yonetimi, workspace, paket bagimliliklari
`system-architect`	Sistem mimarisi, tasarim desenleri, olceklenebilirlik

Uh oh!

FilesExpand file tree

DECKENT.md

Latest commit

History