@hongmaple0820/scale-engine 0.40.2 → 0.44.0

package/README.md

@@ -1,9 +1,9 @@
 <p align="center">
-  <img src="https://img.shields.io/badge/version-0.38.0-orange?style=flat-square" alt="version" />
+  <img src="https://img.shields.io/badge/version-0.42.0-orange?style=flat-square" alt="version" />
   <img src="https://img.shields.io/badge/platforms-22-blue?style=flat-square" alt="platforms" />
   <img src="https://img.shields.io/badge/agents-12-blue?style=flat-square" alt="agents" />
   <img src="https://img.shields.io/badge/tests-verified-brightgreen?style=flat-square" alt="tests" />
-  <img src="https://img.shields.io/badge/npm-0.38.0-cb3837?style=flat-square&logo=npm" alt="npm" />
+  <img src="https://img.shields.io/badge/npm-0.42.0-cb3837?style=flat-square&logo=npm" alt="npm" />
 </p>
 
 [![RepoStars](https://repostars.dev/api/embed?repo=hongmaple0820%2Fscale-engine&theme=copper)](https://repostars.dev/?repos=hongmaple0820%2Fscale-engine&theme=copper)
@@ -114,6 +114,34 @@ scale doctor env --json
 | **Role Skills** | 6 个角色化审查视角（eng-manager、security-reviewer、qa-lead 等） |
 | **Memory Intelligence** | 6 信号质量评分，跨 provider 冲突检测，新鲜度衰减 |
 | **Governance ROI** | 端到端治理 ROI 度量 — token 成本 vs 质量 vs 门禁摩擦 |
+| **Scale Shield** | 退出码钩子拦截引擎 — YAML 策略 → hook 脚本编译，40+ 危险命令阻断，退出码协议 |
+| **Scale Orchestrator** | 声明式编排守护进程 — SCALE_POLICY.md 策略驱动，git worktree 隔离，协调循环 |
+| **Scale Cortex** | 证据驱动持续进化 — 本能提取 (Instincts 0.3-0.9)，SessionStart 注入，跨 harness 适配器 |
+
+## SCALE 2.0 三引擎架构
+
+SCALE 2.0 引入三层引擎，对齐业界前沿项目的核心模式：
+
+| 引擎 | 对标项目 | 核心能力 |
+|------|---------|---------|
+| **Scale Shield** | agent-hooks-in-depth | 退出码阻断 (exit 0/2)、stdin/stdout JSON 协议、40+ 危险命令拦截、`.scale/` 完整性保护 |
+| **Scale Orchestrator** | Symphony WORKFLOW.md | 声明式策略驱动、git worktree 隔离、协调循环、多轮 Worker |
+| **Scale Cortex** | ECC Instincts | 观察→模式→本能提取 (置信度 0.3-0.9)、SessionStart 注入、跨 harness (Claude/Codex/Cursor/Gemini) |
+
+```bash
+# Shield: 编译策略并安装 hook
+scale shield compile
+scale shield status
+
+# Orchestrator: 启动声明式编排守护进程
+scale orch start
+scale orch status
+
+# Cortex: 从失败中学习
+scale cortex evolve
+scale cortex metrics --days 30
+scale cortex inject --minimal
+```
 
 ## AI OS Runtime
 

package/dist/api/cli.js

@@ -17,7 +17,7 @@ import { ProjectAnatomy } from '../context/ProjectAnatomy.js';
 import { buildContextPack, doctorContextBudget, scanContextBudget, writeContextBudgetReport, } from '../context/ContextBudget.js';
 import { resolvePromptCachePolicy } from '../routing/PromptCachePolicy.js';
 import { CerebrumManager } from '../knowledge/CerebrumManager.js';
-import { buildCodeGraphContext, createCodeGraphRoiReport, impactCodeGraph, inspectCodeIntelligence, queryCodeGraph, writeCodeIntelligenceConfig, } from '../codegraph/CodeIntelligence.js';
+import { buildCodeGraphContext, createCodeGraphRoiReport, dumpCodeGraphData, impactCodeGraph, inspectCodeIntelligence, queryCodeGraph, writeCodeIntelligenceConfig, } from '../codegraph/CodeIntelligence.js';
 import { WorkflowEvalStore, compareWorkflowEvalRuns, renderWorkflowEvalReport, runWorkflowEvalSuite, } from '../eval/WorkflowEval.js';
 import { FSMAgentBridge } from '../fsm/FSMAgentBridge.js';
 import { CapabilityRegistry } from '../capabilities/CapabilityRegistry.js';
@@ -50,6 +50,15 @@ import { preflightGateStages } from '../workflow/GateCatalog.js';
 import { gatesCommand } from '../cli/gateStatusCommands.js';
 import { scoreCommand } from '../cli/scoreCommands.js';
 import { promptCommand } from '../cli/promptCommands.js';
+import { quickstartCommand } from '../cli/quickstartCommands.js';
+import { tuiCommand } from '../cli/tuiCommands.js';
+import { qaCommand } from '../cli/qaCommands.js';
+import { autofixCommand } from '../cli/autofixCommands.js';
+import { costReportCommand, costOptimizeCommand } from '../cli/costCommands.js';
+import { reviewCommand as crossReviewCommand } from '../review/reviewCommands.js';
+import { shieldCommand } from '../cli/shieldCommands.js';
+import { orchCommand } from '../cli/orchCommands.js';
+import { cortexCommand } from '../cli/cortexCommands.js';
 import { writeGovernanceTemplates } from '../workflow/GovernanceTemplates.js';
 import { getBootstrapPlanForProfile, getProfile as getConfigProfile, generateConfigForProfile, listProfiles as listConfigProfiles, } from '../config/profiles.js';
 import { computeGovernanceDrift } from '../workflow/GovernanceLock.js';
@@ -1371,9 +1380,31 @@ function printCodeGraphReport(report) {
     for (const warning of report.warnings)
         console.log(`  warning: ${warning}`);
 }
+const codegraphDump = defineCommand({
+    meta: { name: 'dump', description: 'Dump full topology graph (nodes + edges) for visualization' },
+    args: {
+        dir: { type: 'string', default: PROJECT_DIR, description: 'Project directory' },
+        out: { type: 'string', description: 'Output file path (default: stdout as JSON)' },
+    },
+    run({ args }) {
+        const projectDir = resolve(String(args.dir ?? PROJECT_DIR));
+        const scaleDir = resolveScaleDirForProject(projectDir);
+        const graph = dumpCodeGraphData({ projectDir, scaleDir });
+        const json = JSON.stringify(graph, null, 2);
+        const outPath = args.out ? resolve(String(args.out)) : undefined;
+        if (outPath) {
+            mkdirSync(dirname(outPath), { recursive: true });
+            writeFileSync(outPath, json, 'utf-8');
+            console.log(`Topology written to ${outPath} (${graph.nodes.length} nodes, ${graph.edges.length} edges)`);
+        }
+        else {
+            console.log(json);
+        }
+    },
+});
 const codegraph = defineCommand({
     meta: { name: 'codegraph', description: 'Adapter-first code intelligence and exploration ROI' },
-    subCommands: { status: codegraphStatus, init: codegraphInit, query: codegraphQuery, impact: codegraphImpact, context: codegraphContext, roi: codegraphRoi },
+    subCommands: { status: codegraphStatus, init: codegraphInit, query: codegraphQuery, impact: codegraphImpact, context: codegraphContext, roi: codegraphRoi, dump: codegraphDump },
 });
 // ============================================================================
 // eval command - Workflow eval baseline and failure replay
@@ -6548,6 +6579,16 @@ const main = defineCommand({
         'create-prd': createPRD,
         'out-of-scope': outOfScope,
         config,
+        quickstart: quickstartCommand,
+        tui: tuiCommand,
+        qa: qaCommand,
+        'auto-fix': autofixCommand,
+        'cost-report': costReportCommand,
+        'cost-optimize': costOptimizeCommand,
+        'cross-review': crossReviewCommand,
+        shield: shieldCommand,
+        orch: orchCommand,
+        cortex: cortexCommand,
     },
 });
 normalizeUpgradeRootOptionValues(process.argv);