chati-dev 2.0.3 → 2.0.4

package/assets/logo.svg

@@ -35,8 +35,8 @@
   <ellipse cx="77" cy="100" rx="3.5" ry="2.5" fill="rgba(255,255,255,0.3)"/>
   <ellipse cx="133" cy="100" rx="3.5" ry="2.5" fill="rgba(255,255,255,0.3)"/>
 
-  <!-- "chati.dev" text inline -->
+  <!-- "Chati.dev" text inline -->
   <text x="210" y="122" font-family="ui-monospace,'SF Mono','Menlo','Consolas','Liberation Mono',monospace" font-weight="700" font-size="68" letter-spacing="-1">
-    <tspan fill="#e6edf3">chati</tspan><tspan fill="#4A90D9">.</tspan><tspan fill="#e6edf3">dev</tspan>
+    <tspan fill="#e6edf3">Chati</tspan><tspan fill="#4A90D9">.</tspan><tspan fill="#e6edf3">dev</tspan>
   </text>
 </svg>

package/framework/agents/build/dev.md

@@ -338,6 +338,19 @@ Rules:
 
 ---
 
+## Parallelization
+
+```
+This agent supports TASK-LEVEL parallelization (all modes):
+- Independent tasks (no shared file dependencies) MUST run in parallel terminals
+- Tasks with dependencies run sequentially within their dependency chain
+- Each parallel terminal gets isolated write scope per task
+- Orchestrator monitors all terminals and merges results after each batch
+- See Transition Logic step 4.5, Group 2 for details
+```
+
+---
+
 ## Input
 
 $ARGUMENTS

package/framework/agents/clarity/architect.md

@@ -362,7 +362,7 @@ Beyond self-validation (Protocol 5.1), the Architect agent enforces:
 
 ---
 
-## Parallelization Hints
+## Parallelization
 
 - **Can run in parallel with**: Detail agent and UX agent (all three activate post-Brief in parallel-eligible pipelines)
 - **Cannot run in parallel with**: Brief agent (upstream dependency), Phases agent (downstream dependency — requires architecture as input)

package/framework/agents/clarity/brief.md

@@ -215,7 +215,7 @@ agents:
     score: {calculated}
     criteria_count: 9
     completed_at: "{timestamp}"
-current_agent: detail  # (greenfield) or architect (brownfield)
+next_parallel_group: [detail, architect, ux]  # PARALLEL GROUP — see Transition Logic 4.5
 last_handoff: chati.dev/artifacts/handoffs/brief-handoff.md
 ```
 
@@ -226,7 +226,7 @@ last_handoff: chati.dev/artifacts/handoffs/brief-handoff.md
 **Greenfield:**
 ```
 Next steps:
-1. Continue to Detail/PRD agent (Recommended) — define WHAT we'll build
+1. Continue to Detail + Architect + UX in parallel (Recommended) — 3 agents run simultaneously
 2. Review the Brief report
 3. Adjust Brief content
 ```
@@ -234,7 +234,7 @@ Next steps:
 **Brownfield:**
 ```
 Next steps:
-1. Continue to Architect agent (Recommended) — understand existing constraints first
+1. Continue to Detail + Architect + UX in parallel (Recommended) — 3 agents run simultaneously
 2. Review the Brief report
 3. Adjust Brief content
 ```
@@ -444,7 +444,7 @@ On failure:
 
 ---
 
-## Parallelization Hints
+## Parallelization
 
 ```
 This agent is NOT parallelizable.

package/framework/agents/clarity/brownfield-wu.md

@@ -453,7 +453,7 @@ On failure:
 
 ---
 
-## Parallelization Hints
+## Parallelization
 
 ```
 This agent is NOT parallelizable.

package/framework/agents/clarity/detail.md

@@ -371,7 +371,7 @@ Beyond self-validation (Protocol 5.1), the Detail agent enforces:
 
 ---
 
-## Parallelization Hints
+## Parallelization
 
 - **Can run in parallel with**: Architect agent and UX agent (all three activate post-Brief in parallel-eligible pipelines)
 - **Cannot run in parallel with**: Brief agent (upstream dependency), Phases agent (downstream dependency — requires PRD as input)

package/framework/agents/clarity/greenfield-wu.md

@@ -370,7 +370,7 @@ On failure:
 
 ---
 
-## Parallelization Hints
+## Parallelization
 
 ```
 This agent is NOT parallelizable.

package/framework/agents/clarity/ux.md

@@ -392,7 +392,7 @@ Beyond self-validation (Protocol 5.1), the UX agent enforces:
 
 ---
 
-## Parallelization Hints
+## Parallelization
 
 - **Can run in parallel with**: Detail agent and Architect agent (all three activate post-Brief in parallel-eligible pipelines)
 - **Cannot run in parallel with**: Brief agent (upstream dependency), Phases agent (downstream dependency — requires UX specification as input)

package/framework/orchestrator/chati.md

@@ -164,6 +164,34 @@ When an agent completes (score >= 95%):
   2. Agent updates session.yaml (status: completed, score, completed_at)
   3. Agent updates CLAUDE.md with current state
   4. Orchestrator identifies next agent from pipeline
+
+  ╔══════════════════════════════════════════════════════════════╗
+  ║ 4.5. PARALLELIZATION CHECK (MANDATORY — DO NOT SKIP)       ║
+  ║                                                             ║
+  ║ Before activating next agent, check parallel eligibility:   ║
+  ║                                                             ║
+  ║ GROUP 1 — Clarity Phase (after Brief completes):            ║
+  ║   Agents: [detail, architect, ux]                           ║
+  ║   AUTONOMOUS: Spawn all 3 simultaneously (default)          ║
+  ║   HUMAN-IN-THE-LOOP: Offer parallel option (recommended)    ║
+  ║   Write scopes:                                             ║
+  ║     detail → chati.dev/artifacts/2-PRD/                     ║
+  ║     architect → chati.dev/artifacts/3-Architecture/         ║
+  ║     ux → chati.dev/artifacts/4-UX/                          ║
+  ║   → Merge ALL handoffs before proceeding to Phases          ║
+  ║                                                             ║
+  ║ GROUP 2 — Build Phase (Dev agent tasks):                    ║
+  ║   ALL MODES: Independent tasks ALWAYS run in parallel       ║
+  ║   → Analyze task dependency graph from tasks.md             ║
+  ║   → Tasks with no shared file deps = parallel               ║
+  ║   → Tasks with deps = sequential within their chain         ║
+  ║   → Each terminal gets isolated write scope per task        ║
+  ║   → Merge results after each parallel batch                 ║
+  ║                                                             ║
+  ║ If next agent is NOT in a parallel group:                   ║
+  ║   → Continue with sequential activation (step 5)            ║
+  ╚══════════════════════════════════════════════════════════════╝
+
   5. Update session.yaml: current_agent = next_agent
   6. Update project.state if crossing macro-phase boundary:
      - WU through QA-Planning = clarity
@@ -862,6 +890,7 @@ Level 4 - Graceful Degradation:
 ### Autonomous Mode
 ```
 - Execute pipeline transitions silently when gates pass (score >= 95%)
+- Spawn parallel agent groups automatically (Detail + Architect + UX after Brief; independent Dev tasks in parallel)
 - Auto-resolve simple deviations (redirect to correct agent)
 - Pause only on:
   - Quality gate failure (score < 95%)
@@ -886,25 +915,28 @@ Present suggestion with reasoning. User always has final say.
 
 ---
 
-## Parallelization Hints
+## Parallelization Rules
 
 ### Parallelizable Agent Groups
 ```
-Group 1 (post-Brief):
+Group 1 (post-Brief) — MANDATORY PARALLEL:
   - Detail + Architect + UX
-  - Can run in 3 parallel terminals
+  - MUST run in 3 parallel terminals (autonomous mode)
+  - SHOULD offer parallel option (human-in-the-loop mode)
   - Each writes to isolated artifact directories
-  - Orchestrator collects and merges results
+  - Orchestrator MUST collect and merge ALL handoffs before proceeding to Phases
 
-Group 2 (Dev tasks):
-  - Independent dev tasks can run in N parallel terminals
-  - Each terminal has isolated write scope
+Group 2 (Dev tasks) — MANDATORY PARALLEL:
+  - Independent dev tasks MUST run in N parallel terminals (all modes)
+  - Each terminal has isolated write scope per task
   - Orchestrator monitors and collects results
+  - Tasks with no shared file dependencies run simultaneously
+  - Tasks with dependencies run sequentially within their chain
 
-NOT parallelizable:
+NOT parallelizable (always sequential):
   - WU (needs user interaction)
   - Brief (needs user interaction)
-  - Phases (depends on Detail + Architect + UX)
+  - Phases (depends on Detail + Architect + UX results)
   - Tasks (depends on Phases)
   - QA-Planning (validates everything)
   - QA-Implementation (validates everything)

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "chati-dev",
-  "version": "2.0.3",
+  "version": "2.0.4",
   "description": "AI-Powered Multi-Agent Orchestration System — 13 agents, 6 IDEs, 4 languages",
   "type": "module",
   "bin": {
@@ -45,12 +45,12 @@
   "license": "MIT",
   "repository": {
     "type": "git",
-    "url": "https://github.com/ogabrielalonso/chati-dev.git",
+    "url": "https://github.com/ogabrielalonso/Chati.dev.git",
     "directory": "packages/chati-dev"
   },
-  "homepage": "https://github.com/ogabrielalonso/chati-dev#readme",
+  "homepage": "https://github.com/ogabrielalonso/Chati.dev#readme",
   "bugs": {
-    "url": "https://github.com/ogabrielalonso/chati-dev/issues"
+    "url": "https://github.com/ogabrielalonso/Chati.dev/issues"
   },
   "dependencies": {
     "@clack/prompts": "^0.7.0",