@hongmaple0820/scale-engine 0.27.0 → 0.28.0

package/docs/AI_ENGINEERING_OS_POSITIONING.md

@@ -280,18 +280,20 @@ The promotion step must remain evidence-backed. Automatically generating rules w
 
 ## 6. Roadmap Direction
 
-### 6.1 Immediate Patch: 0.26.1
+### 6.1 Planning Principle
 
-Goal: publish the security patch before expanding strategic scope.
+The roadmap has release horizons plus a long-range vision:
 
-Primary outcomes:
-
-- remove `verify-task` shell execution risk from the published package
-- document safe verification command semantics
-- pin and override flagged dependency versions where applicable
-- preserve production dependency audit health
+| Horizon | Purpose | Claim boundary |
+| --- | --- | --- |
+| 0.27.x baseline | establish the AI OS Runtime primitives and adoption path | "runtime baseline", not "complete AI OS" |
+| 0.28.0 closure | make planning, execution, verification, dashboard, benchmark, and adoption usable as a closed loop | "usable closed-loop beta", not "stable final OS" |
+| 0.29.0 intelligence | make memory, context, and skill routing measurably smarter | "intelligence beta", not proven long-term cognition |
+| 0.30.0 governance maturity | strengthen enterprise governance, upgrade, evaluator, and evolution controls | "governance maturity", not commercial stability |
+| 1.0.0 beta | integrate the loop into a public AI Engineering OS beta | "public beta", backed by demos and benchmark evidence |
+| Long-range vision | keep SCALE moving toward an AI Engineering OS with memory, context, governance, and tool intelligence | directional until backed by eval data |
 
-This is a trust-maintenance release. It should not be mixed with large roadmap changes.
+The near-term work should be aggressive, but public wording must stay precise. SCALE can ship beta capabilities quickly; it should only claim stable, industry-leading AI OS behavior after repeated project evidence, benchmarks, and upgrade validation.
 
 ### 6.2 0.27.0: Cognitive Runtime Layer
 
@@ -334,10 +336,105 @@ Exit criteria:
 - skill recommendations include why, when, and required proof: baseline implemented by skill execution plans
 - context pack generation reports token budget and omissions: baseline implemented by `context.pack.compiler`
 
-### 6.3 0.28.0: Adaptive Governance
+### 6.3 0.27.x: Runtime Baseline and Adoption Path
+
+Theme: make the AI OS Runtime installable, inspectable, and safe to adopt.
+
+Current landing status:
+
+- `scale ai-os plan` exists as the unified planning entry point for governance, context, memory, skill routing, adaptive workflow, and ROI.
+- `scale ai-os run --dry-run` exists as the first beta execution slice.
+- `scale ai-os run --mode guarded --verify "<command>"` executes explicit verification commands through the safe command runner, records each command as runtime evidence, and blocks the run when verification fails.
+- `scale ai-os status --lang zh|en` checks runtime directories, plan/run evidence, guarded verification, dashboard health, benchmark evidence, and adoption evidence in one closed-loop readiness report; when verification evidence is missing, it recommends concrete guarded verification commands from `.scale/verification.json` or `package.json`.
+- `scale ai-os dashboard` summarizes persisted run reports into ready/blocked counts, guarded verification health, pending evidence, failure learning candidates, and next recommendations.
+- `scale ai-os benchmark` runs fixed beta scenarios and reports context token use, estimated savings, memory recall, skill steps, governance modes, and the current dashboard health snapshot.
+- `scale ai-os migrate` creates or verifies the `.scale/ai-os` runtime directories and writes an idempotent migration report.
+- `scale ai-os adopt` runs migrate, the first dry-run, benchmark, and doctor as one adoption path, then writes `.scale/ai-os/adoption.json`.
+- `scale ai-os doctor --lang zh|en` checks AI OS runtime readiness without mutating the project and blocks adoption when required directories or dashboard health are broken.
+- `scale upgrade check/plan` includes AI OS readiness, so existing projects see adoption, migration, and doctor steps through the normal upgrade workflow.
+- The upgrade and adoption CLI surfaces now have human-facing Chinese and English output while preserving JSON for scripts, CI, and agent integrations.
+
+Boundary:
+
+- 0.27.x is the baseline. It proves the runtime surface and adoption path, but it does not yet prove autonomous source mutation, PR creation, long-term memory, or stable commercial AI OS behavior.
+
+### 6.4 0.28.0: Usable Closed-Loop Enhancement
+
+Theme: turn `ai-os plan` into a runnable beta loop.
+
+Target timebox: 2-3 weeks.
+
+Core work:
+
+| Module | Outcome |
+| --- | --- |
+| `scale ai-os run` | execute the unified plan through workflow, context, memory, skill routing, and verification steps |
+| Runtime Status | show whether plan, run, verification, dashboard, benchmark, adoption, and doctor evidence exist for the project |
+| Verification Recommendation | derive suggested verification commands from task level, changed files, project verification profile, and risk signals |
+| Failure Learning Closure | convert failed guarded runs, gate failures, and missing evidence into reviewed lesson/rule candidates |
+| Closed-Loop Demo Pack | provide repeatable docs and code task demos that exercise plan -> run -> verify -> dashboard -> benchmark |
+| Memory Provider Bridge | keep gbrain, agentmemory, code memory, and local memory selectable through one provider contract |
+| Context Compiler v2 | merge task intent, risk level, files, memory recall, and role into one explainable context pack |
+| Skill Router v2 | create an execution graph for skills, MCP tools, CLIs, artifacts, and required evidence |
+| Adaptive Workflow Profiles | choose light, standard, or strict gates from risk and changed-file signals |
+| AI OS Dashboard CLI | summarize gate health, memory hits, context budget, skill evidence, and ROI |
+| Upgrade/Migration | migrate older `.scale` state and warn about incompatible local governance files |
+| AI OS Adoption and Doctor | keep one-command adoption and readiness checks aligned with the normal upgrade workflow |
+| Bilingual DX | keep key CLI help, errors, README guidance, and tutorials readable in Chinese and English |
+| Benchmark Pack | run fixed samples for token budget, recall, gate pass rate, and skill-routing evidence |
+
+Exit criteria:
+
+- `scale ai-os run` can complete at least one documentation task and one code task in dry-run or guarded execution mode
+- `scale ai-os status` or equivalent doctor output shows what is missing for a closed loop
+- verification recommendations are explainable and can be overridden by explicit `--verify` commands
+- execution output records context decisions, memory provider choices, skill decisions, gate results, and failure lessons
+- benchmark output compares context token budget against a full-load baseline
+- beta docs clearly state what is automated, what is proposed, and what still requires human approval
+
+Current implementation status:
+
+- In progress on the post-0.27.1 development branch.
+- Runtime baseline, status visibility, verification recommendation, adoption, doctor, dashboard, benchmark, migration, upgrade integration, and bilingual adoption guidance are already landed.
+- Remaining 0.28.0 work should focus on failure-learning closure and repeatable end-to-end demo evidence.
+- It does not yet create PRs or mutate source files; richer skill execution remains a later implementation slice unless explicitly approved.
+
+Explicitly deferred:
+
+- default automatic PR creation or merge without review
+- deep dynamic dependency sandboxing beyond audit, lockfile diff, and high-risk pattern checks
+- full VLM visual judgment beyond screenshot capture and interface placeholders
+- claims of human-level long-term memory or fully autonomous engineering
+
+### 6.5 0.29.0: Memory, Context, and Skill Intelligence
+
+Theme: make the beta loop measurably smarter rather than only broader.
+
+Target timebox: 4-6 weeks.
+
+Core work:
+
+| Module | Outcome |
+| --- | --- |
+| Memory Quality Scoring | score recall precision, contradiction risk, accepted memory rate, and stale-memory risk |
+| Provider Fallback Policy | choose between gbrain, agentmemory, code memory, local memory, or no memory with an explicit reason |
+| Context Compression | summarize low-risk context while preserving high-risk evidence verbatim |
+| Skill Strategy Learning | learn preferred tools from successful evidence, failures, and user overrides |
+| Workflow Eval Integration | turn benchmark results into release-gate evidence |
+
+Exit criteria:
+
+- memory recall has acceptance/rejection feedback
+- context packs show savings, omissions, and evidence-loss warnings
+- skill routing decisions can be compared against outcome quality
+- release notes include measured deltas instead of aspirational percentages
+
+### 6.6 0.30.0: Enterprise Governance and Upgrade Maturity
 
 Theme: deepen adaptive governance beyond the v0.27.0 baseline.
 
+Target timebox: 6-10 weeks.
+
 Core work:
 
 | Module | Outcome |
@@ -354,10 +451,12 @@ Exit criteria:
 - reasoning-heavy tasks get critique/evaluator gates
 - evolution proposals can be traced to failure evidence and validation results
 
-### 6.4 0.29.0+: Agent Engineering OS
+### 6.7 1.0.0 Beta: AI Engineering OS
 
 Theme: integrate governance, memory, context, and tools into an operating layer.
 
+Target timebox: 8-12 weeks.
+
 Target capabilities:
 
 - unified agent workspace policy
@@ -367,6 +466,37 @@ Target capabilities:
 - measurable token and quality reports
 - ecosystem-safe skill and MCP lifecycle governance
 
+Release criteria:
+
+- install, upgrade, run, dashboard, benchmark, and migration flows work on clean projects
+- at least three representative project types have documented smoke results
+- failure learning produces reviewed rule candidates without silently hardening bad rules
+- bilingual docs explain the core workflow without requiring maintainer context
+- public claims are tied to `WORKFLOW_EVAL`, benchmark output, or release evidence
+
+### 6.8 1.0.0 Stable and Long-Range Vision
+
+This is the strategic north star, not the 0.28.0 closed-loop promise.
+
+| Time horizon | Target state | Evidence required before public claim |
+| --- | --- | --- |
+| 8-12 weeks | AI Engineering OS beta: usable end-to-end loop across planning, execution, verification, memory, and dashboard | repeatable demo projects and benchmark reports |
+| 3-6 months | stable governance runtime: upgrades, adapters, memory providers, and eval gates are reliable in real repositories | release-to-release regression data and field reports |
+| 6-12 months | industry-leading agent engineering layer: adaptive workflows, strategy memory, tool intelligence, and cross-agent governance mature together | comparative evals, sustained issue closure, external adoption evidence |
+
+Long-range capability themes:
+
+- Cognitive memory: working, episodic, semantic, procedural, and strategy memory with explicit source and freshness controls.
+- Adaptive orchestration: workflows selected by risk, ownership, failure history, and tool reliability instead of one fixed path.
+- Tool intelligence: skills, MCP, CLIs, browser automation, and agent adapters treated as governed capabilities with cost, evidence, and fallback policy.
+- Evaluator intelligence: critique loops, uncertainty scoring, adversarial review, and evidence insufficiency verdicts for reasoning-heavy tasks.
+- Governance economics: token cost, gate friction, verification quality, and maintenance overhead measured as first-class product metrics.
+- Ecosystem governance: external skills, memory providers, adapters, and templates integrated through attribution, license, source pinning, and supply-chain checks.
+
+Non-negotiable boundary:
+
+> The long-range vision can guide architecture, but it must not be used as a release claim until the corresponding evidence exists.
+
 ## 7. Measurement Plan
 
 Strategic claims must be tied to measurement.

package/docs/README.md

@@ -36,7 +36,7 @@
 | [CODE_INTELLIGENCE.md](CODE_INTELLIGENCE.md) | CodeGraph、Graphify 和显式 fallback 的代码智能与探索 ROI |
 | [WORKFLOW_EVAL.md](WORKFLOW_EVAL.md) | Workflow Eval、pass@k 指标、Failure Replay 和改进候选 |
 | [SKILL_RADAR.md](SKILL_RADAR.md) | Skill Radar、能力置信度、证据要求和供应链安全检查 |
-| [AI_ENGINEERING_OS_POSITIONING.md](AI_ENGINEERING_OS_POSITIONING.md) | Agent Governance Runtime / AI Engineering OS 方向，以及 `scale ai-os plan` 一体化 runtime plan |
+| [AI_ENGINEERING_OS_POSITIONING.md](AI_ENGINEERING_OS_POSITIONING.md) | Agent Governance Runtime / AI Engineering OS 方向、`scale ai-os plan/run/status/dashboard/benchmark/migrate/adopt/doctor` runtime 入口、`0.28.0` 可用闭环增强和 3-12 个月远景路线 |
 | [THIRD_PARTY_SKILLS.md](THIRD_PARTY_SKILLS.md) | 第三方 skill 致谢、授权边界、引用方式和 vendoring 策略 |
 | [EXTERNAL_REFERENCES.md](EXTERNAL_REFERENCES.md) | 外部项目、skills、MCP、CLI 和适配器引用的完整清单 |
 | [UPGRADE_MANAGEMENT.md](UPGRADE_MANAGEMENT.md) | SCALE CLI、governance pack、skills、MCP 和 CLI 工具的安全升级流程 |

package/docs/start/README.md

@@ -71,8 +71,9 @@ scale status
 | 多仓库/MOE 工作区 | `scale init --governance-pack moe-workspace` |
 | 文档、报告、截图、脚本混乱 | `scale init --governance-pack resource-governance` |
 | 工作流或第三方能力要升级 | `scale upgrade check --lang zh && scale upgrade plan --html --lang zh` |
+| 已有项目接入 AI OS runtime | `scale ai-os adopt --task "接入 AI OS runtime" --lang zh` |
 
 
 ## 工作流升级短路径
 
-已有项目先看 [SCALE 工作流升级指南](workflow-upgrade.md)。它说明 `scale init --interactive`、`scale upgrade check/plan/apply/rollback`、`--lang zh/en` 双语输出、仓库本地 `make workflow-upgrade-*` 入口，以及生成文件更新和项目级验证之间的边界。
+已有项目先看 [SCALE 工作流升级指南](workflow-upgrade.md)。它说明 `scale init --interactive`、`scale upgrade check/plan/apply/rollback`、`scale ai-os adopt`、`--lang zh/en` 双语输出、仓库本地 `make workflow-upgrade-*` / `make workflow-aios-adopt` 入口，以及生成文件更新和项目级验证之间的边界。

package/docs/start/quickstart.md

@@ -131,6 +131,12 @@ scale upgrade plan --dir . --html --lang zh
 scale upgrade apply --dir . --confirm --lang zh
 ```
 
+如果升级计划提示 AI OS runtime 尚未接入，用一键接入命令生成运行态目录、首份 dry-run、benchmark 和 doctor 报告：
+
+```bash
+scale ai-os adopt --dir . --task "接入 AI OS runtime" --lang zh
+```
+
 需要英文输出时把 `--lang zh` 换成 `--lang en`。干净的 SCALE 受管文件可以自动刷新；已有本地改动的文件会进入人工审阅，不会被自动覆盖。
 
 继续阅读 [官方 Demo Walkthrough](agent-governance-demo.md)，看一个真实任务如何从需求到验证证据。

package/docs/start/workflow-upgrade.md

@@ -60,6 +60,18 @@ scale upgrade apply --dir . --confirm
 scale preflight --dir . --service all --preflight-profile quick
 ```
 
+如果升级计划提示 AI OS runtime 尚未接入，优先使用一键接入命令。它会创建运行态目录、生成首个 `dry-run` 运行报告、写入 benchmark，并用 doctor 复核就绪状态：
+
+```bash
+scale ai-os adopt \
+  --dir . \
+  --task "接入 AI OS runtime 并生成首份治理证据" \
+  --files "README.md,AGENTS.md" \
+  --lang zh
+```
+
+接入完成后会写入 `.scale/ai-os/adoption.json`。后续真实任务再使用 `scale ai-os run --mode guarded` 生成受治理的执行证据。
+
 默认输出是中文。需要英文命令提示或英文 HTML 计划时加 `--lang en`：
 
 ```bash
@@ -67,12 +79,15 @@ scale upgrade check --dir . --lang en
 scale upgrade plan --dir . --html --lang en
 ```
 
+给人看的升级输出会使用当前语言生成下一步命令，例如中文场景会推荐 `scale ai-os adopt --task "接入 AI OS runtime" --lang zh`。只有脚本、CI 或 Agent 集成需要稳定结构时才使用 `--json`。
+
 如果仓库已有本地封装，优先使用本地命令，因为它们编码了项目默认值：
 
 ```bash
 make workflow-upgrade-check
 make workflow-upgrade-plan
 make workflow-upgrade-apply
+make workflow-aios-adopt
 make workflow-upgrade-verify
 ```
 
@@ -160,6 +175,8 @@ workflow-upgrade-rollback:
 	scale upgrade rollback --dir . --lang zh
 workflow-upgrade-verify:
 	scale preflight --dir . --service all --preflight-profile quick
+workflow-aios-adopt:
+	scale ai-os adopt --dir . --task "$(TASK)" --files "$(FILES)" --level "$(LEVEL)" --budget "$(BUDGET)" --lang zh
 ```
 
 如果 Windows 环境没有 `make`，提供等价 PowerShell 脚本，或在文档里写清原始 `scale` 命令。

package/docs/workflow/README.md

@@ -62,6 +62,7 @@ feature/fix/docs/chore/codex -> dev -> master
 make bootstrap-scale
 make workflow-upgrade-check
 make workflow-upgrade-plan
+make workflow-aios-adopt
 ```
 
-先审计划，再决定是否 `make workflow-upgrade-apply`。
+先审计划，再决定是否 `make workflow-upgrade-apply`。如果计划提示 AI OS runtime 尚未接入，使用 `make workflow-aios-adopt` 生成运行态目录、首份 dry-run、benchmark 和 doctor 报告。

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@hongmaple0820/scale-engine",
-  "version": "0.27.0",
+  "version": "0.28.0",
   "description": "Executable AI agent governance with workflow gates, evidence, skill/tool orchestration, and traceable HTML artifacts",
   "repository": {
     "type": "git",