Phase 10: Goal Decomposer — LLM-powered goal → task DAG → parallel execution (v4.15.0)

Added:
- GoalDecomposer: LLM-powered goal decomposition into validated TaskDAG
- TeamRunner: DAG execution engine with topological-layer scheduling,
  concurrency limits, per-task/total timeouts, dependency result injection
- runTeam() one-liner: single function call from goal to results
- createLLMPlanner(): structured prompt → any LLM via adapter system
- DAG utilities: validateDAG (Kahn's cycle detection), topologicalLayers,
  parsePlanJSON (robust LLM response parsing)
- TypeScript 6.0 upgrade (5.9.3 → 6.0.2) with ignoreDeprecations: 6.0
- postinstall script for third-party tsconfig patching
- 153 new tests in test-phase10.ts (2,357 total across 25 suites)

New module: lib/goal-decomposer.ts
Exports: GoalDecomposer, TeamRunner, runTeam, createLLMPlanner,
  validateDAG, topologicalLayers, parsePlanJSON + all types

Author: jovanSAPFIONEER

.github/SECURITY.md

@@ -4,7 +4,9 @@
 
 | Version | Supported |
 |---------|-----------|
-| 4.13.x  | ✅ Yes — full support (current) |
+| 4.15.x  | ✅ Yes — full support (current) |
+| 4.14.x  | ✅ Security fixes only |
+| 4.13.x  | ✅ Security fixes only |
 | 4.12.x  | ✅ Security fixes only |
 | 4.9.x   | ✅ Security fixes only |
 | 4.8.x   | ✅ Security fixes only |
@@ -51,6 +53,8 @@ Network-AI includes built-in security features:
 - **Approval Gates** (v4.14.0) -- `ApprovalGate` requires explicit human or callback approval for high-risk operations (writes, shell commands, budget spend); auto-approve mode for trusted environments; full approval history with audit trail
 - **Pipe Mode Authentication** (v4.14.0) -- JSON stdin/stdout protocol for programmatic agent control; commands processed one-at-a-time with structured responses; no shell injection surface
 - **Strategy Agent Pool Isolation** (v4.14.0) -- `AgentPool` enforces per-pool capacity ceilings; `WorkloadPartitioner` routes tasks by priority class; adaptive scaling respects budget constraints before spawning agents
+- **Goal Decomposer DAG Validation** (v4.15.0) -- `validateDAG()` enforces acyclicity (Kahn's algorithm), rejects self-dependencies and unknown task references; task graphs are validated before execution to prevent infinite loops or orphaned tasks
+- **Team Runner Approval Gate** (v4.15.0) -- optional `approvalCallback` on `runTeam()` requires explicit approval of the full task DAG before any agent execution begins; rejection skips all tasks with audit-ready status
 
 ## Security Scan Results
 

.github/copilot-instructions.md

@@ -2,7 +2,7 @@
 
 ## Project Overview
 
-Network-AI is a TypeScript/Node.js multi-agent orchestrator — shared state, guardrails, budgets, and cross-framework coordination (v4.14.0). 2,204 tests across 24 suites.
+Network-AI is a TypeScript/Node.js multi-agent orchestrator — shared state, guardrails, budgets, and cross-framework coordination (v4.15.0). 2,357 tests across 25 suites.
 
 ## Architecture
 
@@ -36,13 +36,14 @@ Network-AI is a TypeScript/Node.js multi-agent orchestrator — shared state, gu
 - `lib/agent-runtime.ts` — AgentRuntime sandboxed execution with SandboxPolicy, ShellExecutor, FileAccessor, ApprovalGate
 - `lib/console-ui.ts` — ConsoleUI interactive terminal dashboard
 - `lib/strategy-agent.ts` — StrategyAgent meta-orchestrator with AgentPool, WorkloadPartitioner, adaptive scaling
+- `lib/goal-decomposer.ts` — GoalDecomposer, TeamRunner, runTeam: LLM-powered goal → task DAG → parallel execution
 - `adapters/` — 17 framework adapters (LangChain, AutoGen, CrewAI, MCP, Codex, MiniMax, NemoClaw, APS, etc.)
 
 ## Build & Test
 
 ```bash
 npx tsc --noEmit              # Type-check (zero errors expected)
-npm run test:all              # All 2,204 tests across 24 suites
+npm run test:all              # All 2,357 tests across 25 suites
 npm test                      # Core orchestrator tests
 npm run test:adapters         # All 17 adapters
 ```

ARCHITECTURE.md

@@ -303,7 +303,11 @@ Network-AI/
 │   ├── semantic-search.ts        # BYOE vector store with cosine similarity (v4.12)
 │   ├── phase-pipeline.ts         # Multi-phase workflows with approval gates (v4.13)
 │   ├── confidence-filter.ts      # Multi-agent result scoring and filtering (v4.13)
-│   └── fan-out.ts                # Parallel agent spawning with pluggable aggregation (v4.13)
+│   ├── fan-out.ts                # Parallel agent spawning with pluggable aggregation (v4.13)
+│   ├── agent-runtime.ts          # Sandboxed execution with SandboxPolicy, ShellExecutor (v4.14)
+│   ├── console-ui.ts             # Interactive terminal dashboard with ANSI TUI (v4.14)
+│   ├── strategy-agent.ts         # Meta-orchestrator with AgentPool, WorkloadPartitioner (v4.14)
+│   └── goal-decomposer.ts        # LLM-powered goal → task DAG → parallel execution (v4.15)
 ├── scripts/                      # Python helper scripts (local orchestration only)
 │   ├── blackboard.py             # Shared state management with atomic commits
 │   ├── swarm_guard.py            # Handoff tax prevention, budget tracking

CHANGELOG.md

@@ -5,6 +5,17 @@ All notable changes to Network-AI will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [4.15.0] - 2026-04-04
+
+### Added
+- **Goal Decomposer** (`GoalDecomposer`) — LLM-powered goal → task DAG → parallel execution. Takes a natural language goal, decomposes it into a validated `TaskDAG` via an LLM planner, respects dependencies, and executes with concurrency control. New module: `lib/goal-decomposer.ts`.
+- **Team Runner** (`TeamRunner`) — DAG execution engine with topological-layer scheduling, concurrency limits, per-task and total timeouts, dependency result injection (`_dependencyResults`), priority ordering within layers, and `continueOnFailure` mode.
+- **`runTeam()` one-liner** — single function call: `runTeam(goal, agents, { planner, executor })` to go from natural language goal to results. Includes optional approval gate, planner retries, and full event emission.
+- **`createLLMPlanner()`** — built-in planner factory that sends structured prompts to any LLM agent via the adapter system and parses JSON responses (handles code fences, preamble text, nested `{ tasks }` / `{ text }` / `{ content }` shapes).
+- **DAG utilities** — `validateDAG()` (cycle detection via Kahn's algorithm, self-dependency and unknown-ref checks), `topologicalLayers()` (parallel scheduling), `parsePlanJSON()` (robust LLM response parsing).
+- **TypeScript 6.0** — upgraded from 5.9.3 to 6.0.2; added `ignoreDeprecations: "6.0"` and postinstall script for third-party tsconfig patching.
+- 153 new tests in `test-phase10.ts` (2,357 total across 25 suites)
+
 ## [4.14.0] - 2026-04-02
 
 ### Added

CLAUDE.md

@@ -4,14 +4,14 @@ This file is read automatically by Claude Code when working in this repository.
 
 ## Project Overview
 
-Network-AI is a TypeScript/Node.js multi-agent orchestrator — shared state, guardrails, budgets, and cross-framework coordination. Version 4.14.0.
+Network-AI is a TypeScript/Node.js multi-agent orchestrator — shared state, guardrails, budgets, and cross-framework coordination. Version 4.15.0.
 
 ## Build & Test Commands
 
 ```bash
 npm install                   # Install dependencies
 npx tsc --noEmit              # Type-check (zero errors expected)
-npm run test:all              # Run all 2,204 tests across 24 suites
+npm run test:all              # Run all 2,357 tests across 25 suites
 npm test                      # Core orchestrator tests only
 npm run test:security         # Security module tests
 npm run test:adapters         # All 17 adapter tests
@@ -37,6 +37,7 @@ All tests must pass before any commit. No test should be skipped or marked `.onl
 - `lib/agent-runtime.ts` — AgentRuntime: sandboxed execution with SandboxPolicy, ShellExecutor, FileAccessor, ApprovalGate
 - `lib/console-ui.ts` — ConsoleUI: interactive terminal dashboard with ANSI TUI
 - `lib/strategy-agent.ts` — StrategyAgent: meta-orchestrator with AgentPool, WorkloadPartitioner, adaptive scaling
+- `lib/goal-decomposer.ts` — GoalDecomposer, TeamRunner, runTeam: LLM-powered goal → task DAG → parallel execution
 - `adapters/` — 17 framework adapters (LangChain, AutoGen, CrewAI, MCP, Codex, MiniMax, NemoClaw, APS, etc.)
 - `bin/cli.ts` — CLI entry point (`npx network-ai`)
 - `bin/mcp-server.ts` — MCP server (SSE + stdio transport)

CODEX.md

@@ -4,14 +4,14 @@ This file is read automatically by OpenAI Codex CLI when working in this reposit
 
 ## Project Overview
 
-Network-AI is a TypeScript/Node.js multi-agent orchestrator — shared state, guardrails, budgets, and cross-framework coordination. Version 4.14.0.
+Network-AI is a TypeScript/Node.js multi-agent orchestrator — shared state, guardrails, budgets, and cross-framework coordination. Version 4.15.0.
 
 ## Build & Test Commands
 
 ```bash
 npm install                   # Install dependencies
 npx tsc --noEmit              # Type-check (zero errors expected)
-npm run test:all              # Run all 2,204 tests across 24 suites
+npm run test:all              # Run all 2,357 tests across 25 suites
 npm test                      # Core orchestrator tests only
 npm run test:security         # Security module tests
 npm run test:adapters         # All 17 adapter tests
@@ -37,6 +37,7 @@ All tests must pass before any commit. No test should be skipped or marked `.onl
 - `lib/agent-runtime.ts` — AgentRuntime: sandboxed execution with SandboxPolicy, ShellExecutor, FileAccessor, ApprovalGate
 - `lib/console-ui.ts` — ConsoleUI: interactive terminal dashboard with ANSI TUI
 - `lib/strategy-agent.ts` — StrategyAgent: meta-orchestrator with AgentPool, WorkloadPartitioner, adaptive scaling
+- `lib/goal-decomposer.ts` — GoalDecomposer, TeamRunner, runTeam: LLM-powered goal → task DAG → parallel execution
 - `adapters/` — 17 framework adapters (LangChain, AutoGen, CrewAI, MCP, Codex, MiniMax, NemoClaw, APS, etc.)
 - `bin/cli.ts` — CLI entry point (`npx network-ai`)
 - `bin/mcp-server.ts` — MCP server (SSE + stdio transport)

CONTRIBUTING.md

@@ -78,7 +78,7 @@ Unsolicited PRs without a linked, approved issue will be closed.
 git clone https://github.com/Jovancoding/Network-AI.git
 cd Network-AI
 npm install
-npm run test:all  # Run all 2,204 tests (24 suites)
+npm run test:all  # Run all 2,357 tests (25 suites)
 npm run test:phase4  # Phase 4 behavioral control plane tests only
 npx tsc --noEmit  # Type-check
 ```

ENTERPRISE.md

@@ -100,7 +100,7 @@ Network-AI follows [Semantic Versioning](https://semver.org/):
 
 ### Stability Signals
 
-- 2,204 passing assertions across 24 suites
+- 2,357 passing assertions across 25 suites
 - Deterministic scoring — no random outcomes in permission evaluation or budget enforcement
 - CI runs on every push and every PR
 - All examples ship with the repo and run without mocking

INTEGRATION_GUIDE.md

@@ -409,7 +409,7 @@ Run these before declaring the integration production-ready:
 - [ ] `npx ts-node test-phase4.ts` — 147 behavioral tests pass
 - [ ] `npx ts-node test-qa.ts` — 67 QA orchestrator tests pass
 - [ ] `npx ts-node test-phase7.ts` — 94 Phase 7 tests pass (hooks, flow control, composer, semantic search)
-- [ ] `npm run test:all` — all 2,204 tests pass across 24 suites
+- [ ] `npm run test:all` — all 2,357 tests pass across 25 suites
 - [ ] `npm run demo -- --08` runs to completion in < 10 seconds
 
 ### Race Condition Safety
@@ -477,4 +477,4 @@ Run these before declaring the integration production-ready:
 
 ---
 
-*Network-AI v4.14.0 · MIT License · https://github.com/Jovancoding/Network-AI*
+*Network-AI v4.15.0 · MIT License · https://github.com/Jovancoding/Network-AI*

README.md

@@ -5,14 +5,14 @@
 [![Website](https://img.shields.io/badge/website-network--ai.org-4b9df2?style=flat&logo=web&logoColor=white)](https://network-ai.org/)
 [![CI](https://github.com/Jovancoding/Network-AI/actions/workflows/ci.yml/badge.svg)](https://github.com/Jovancoding/Network-AI/actions/workflows/ci.yml)
 [![CodeQL](https://github.com/Jovancoding/Network-AI/actions/workflows/codeql.yml/badge.svg)](https://github.com/Jovancoding/Network-AI/actions/workflows/codeql.yml)
-[![Release](https://img.shields.io/badge/release-v4.14.0-blue.svg)](https://github.com/Jovancoding/Network-AI/releases)
+[![Release](https://img.shields.io/badge/release-v4.15.0-blue.svg)](https://github.com/Jovancoding/Network-AI/releases)
 [![npm](https://img.shields.io/npm/dw/network-ai.svg?label=npm%20downloads)](https://www.npmjs.com/package/network-ai)
-[![Tests](https://img.shields.io/badge/tests-2204%20passing-brightgreen.svg)](#testing)
+[![Tests](https://img.shields.io/badge/tests-2357%20passing-brightgreen.svg)](#testing)
 [![Adapters](https://img.shields.io/badge/frameworks-17%20supported-blueviolet.svg)](#adapter-system)
 [![License](https://img.shields.io/badge/license-MIT-brightgreen.svg)](LICENSE)
 [![Socket](https://socket.dev/api/badge/npm/package/network-ai)](https://socket.dev/npm/package/network-ai/overview)
 [![Node.js](https://img.shields.io/badge/node-%3E%3D18.0.0-brightgreen.svg)](https://nodejs.org)
-[![TypeScript](https://img.shields.io/badge/TypeScript-5.x-3178C6.svg)](https://typescriptlang.org)
+[![TypeScript](https://img.shields.io/badge/TypeScript-6.x-3178C6.svg)](https://typescriptlang.org)
 [![ClawHub](https://img.shields.io/badge/ClawHub-network--ai-orange.svg)](https://clawhub.ai/skills/network-ai)
 [![Integration Guide](https://img.shields.io/badge/docs-integration%20guide-informational.svg)](INTEGRATION_GUIDE.md)
 [![Sponsor](https://img.shields.io/badge/sponsor-support%20the%20project-f15bb5.svg)](https://github.com/sponsors/Jovancoding)
@@ -102,6 +102,7 @@ Runs priority preemption, AuthGuardian permission gating, FSM governance, and co
 | ✅ Interactive console | TUI dashboard for live monitoring, agent control, blackboard/budget/FSM management |
 | ✅ Pipe mode | JSON stdin/stdout protocol for programmatic AI-to-orchestrator control |
 | ✅ Strategy agent | Meta-orchestrator with elastic agent pools, workload partitioning, and adaptive scaling |
+| ✅ Goal decomposer | LLM-powered goal → task DAG → parallel execution with `runTeam()` one-liner |
 | ✅ TypeScript native | ES2022 strict mode, zero native dependencies |
 
 ---
@@ -396,9 +397,10 @@ npm run test:codex        # Codex adapter
 npm run test:priority     # Priority & preemption
 npm run test:cli          # CLI layer
 npm run test:phase9       # Agent runtime, console, strategy agent
+npm run test:phase10      # Goal decomposer, task DAG, runTeam
 ```
 
-**2,204 passing assertions across 24 test suites** (`npm run test:all`):
+**2,357 passing assertions across 25 test suites** (`npm run test:all`):
 
 | Suite | Assertions | Covers |
 |---|---|---|
@@ -425,6 +427,7 @@ npm run test:phase9       # Agent runtime, console, strategy agent
 | `test-phase7.ts` | 94 | Deferred init, hook middleware, flow control, skill composer, semantic search |
 | `test-phase8.ts` | 146 | Phase pipeline, confidence filter, matcher-based hooks, fan-out/fan-in |
 | `test-phase9.ts` | 280 | Agent runtime, sandbox policy, shell executor, file accessor, approval gate, console UI, orchestrator wiring, pipe mode, strategy agent |
+| `test-phase10.ts` | 153 | Goal decomposer, task DAG validation, topological layers, JSON parsing, team runner, concurrency, timeouts, events, runTeam one-liner, dependency injection, LLM planner |
 
 ---