@oevery/rsp 3.1.0-beta.0 → 3.1.0-beta.2

package/README.md

@@ -4,7 +4,7 @@ English | [简体中文](./README.zh-CN.md)
 
 **A repository-native engineering workflow for humans and AI agents.**
 
-RSP helps turn unclear intent into shaped, implemented, reviewed, and verified software changes while keeping project knowledge durable and work recoverable. Its ten composable Skills guide the engineering journey without introducing hidden workflow state or replacing the project's own files, tools, and authority.
+RSP helps turn unclear intent into shaped, implemented, reviewed, and verified software changes while keeping project knowledge durable and work recoverable. Its eleven composable Skills guide the engineering journey without introducing hidden workflow state or replacing the project's own files, tools, and authority.
 
 The workflow rests on a lightweight artifact foundation of **Rules, Specs, and Plans**.
 
@@ -15,7 +15,7 @@ The workflow rests on a lightweight artifact foundation of **Rules, Specs, and P
 - Route unexplained failures to diagnosis, ordinary evidenced edits to implementation, and only explicit or concrete-risk test-first work to TDD.
 - Review changes against explicit scope, address accepted findings, and rerun fresh verification.
 - Preserve stable facts, scoped instructions, lasting rationale, and completed history in their proper owners.
-- Prepare release documentation when a Change explicitly owns a confirmed release.
+- Prepare or reconcile release documentation for an explicit confirmed release operation; use a Release Change only when material coordination needs a persistent owner.
 
 ## How the workflow fits together
 
@@ -25,7 +25,7 @@ intent
   → design when needed
   → diagnose | TDD | implement
   → review → address accepted findings
-  → release docs when owned
+  → release docs for an explicit confirmed operation
   → durable review → archive
 ```
 
@@ -34,17 +34,17 @@ RSP derives the next action from the selected Change, repository evidence, verif
 ## Quick start
 
 ```bash
-npx -y @oevery/rsp@3.1.0-beta.0 init
-npx -y @oevery/rsp@3.1.0-beta.0 doctor
+npx -y @oevery/rsp@3.1.0-beta.2 init
+npx -y @oevery/rsp@3.1.0-beta.2 doctor
 ```
 
 Recommended bootstrap flow:
 
 ```bash
-npx -y @oevery/rsp@3.1.0-beta.0 init --with-project-setup
+npx -y @oevery/rsp@3.1.0-beta.2 init --with-project-setup
 # fill .rsp/changes/project-setup.md
 # fill .rsp/specs/design.md
-npx -y @oevery/rsp@3.1.0-beta.0 doctor
+npx -y @oevery/rsp@3.1.0-beta.2 doctor
 ```
 
 ## Artifact foundation
@@ -105,7 +105,7 @@ npx -y @oevery/rsp@3.1.0-beta.0 doctor
 - `.rsp/specs/design.md`: created by `rsp init`, then owned by the project.
 - `.rsp/specs/decisions/`: the default authoritative Decision Record directory; configure `decisions.path` only when the Host Project already owns one external ADR directory.
 - `.rsp/rsp-rules.md`: generated minimal fallback protocol; use the `rsp` skill when available.
-- Keep durable architecture, boundaries, and cross-cutting technical constraints in `.rsp/specs/design.md`.
+- Keep project-wide boundaries and navigation in `.rsp/specs/design.md`; move cohesive reusable facts to the smallest domain Spec listed by `.rsp/specs/INDEX.md`.
 - Treat `.rsp/specs/INDEX.md` as a directory for additional spec files; it does not list `design.md`.
 - Keep stable scoped workflow and validation instructions in the nearest project-owned `AGENTS.md`, outside the managed RSP block.
 
@@ -149,79 +149,80 @@ Do not treat `.rsp/specs/` or `.rsp/changes/` as replacements for nearest `AGENT
 
 ## Skills
 
-RSP publishes ten host-neutral Skills for on-demand loading:
+RSP publishes eleven default host-neutral lifecycle Skills for on-demand loading, plus one explicitly installed project-level audit Skill:
+
+| Skill | Role |
+|---|---|
+| `rsp` | Derive the next action; guide setup, durable review, and archive decisions. |
+| `rsp-shape` | Shape one executable Change or justified shallow Group without implementation. |
+| `rsp-design` | Resolve one bounded domain, module/seam, or reversible-exploration question before or within tracked work. |
+| `rsp-implement` | Implement one selected ready Change with fresh verification. |
+| `rsp-diagnose` | Confirm a cause, or return a truthful unresolved diagnosis, before correction. |
+| `rsp-tdd` | Drive one clear behavior through RED, GREEN, and safe REFACTOR. |
+| `rsp-review` | Review a fixed code, document, or mixed comparison without mutation. |
+| `rsp-address-review` | Dispose fixed findings, correct accepted ones, verify, and request re-review. |
+| `rsp-commit` | Create one authorized exact-scope local commit with a repository-consistent structured message. |
+| `rsp-release-docs` | Draft, audit, finalize, or reconcile evidence-based release surfaces. |
+| `rsp-manage` | Continue one explicitly requested or project-enabled eligible ready Change or shallow Group. |
+| `rsp-codebase-audit` (optional) | Audit one bounded repository or subtree for evidence-backed structural risks without mutation. |
+
+Each lifecycle Skill returns to an existing project or RSP owner. Report-only Pre-Change Design and `rsp-codebase-audit` may return one bounded result directly to the user without inventing an artifact owner. The suite adds no hidden workflow state or recursive Skill orchestration. No Skill infers commit, push, publication, deployment, approval, or human-acceptance authority.
 
-- `rsp`: setup, workflow, durable review, and archive guidance.
-- `rsp-shape`: shape unclear non-trivial work into one ready Change or justified shallow Group without implementing it; progressively load deep clarification for an explicit rigorous challenge or unresolved high-risk decision, and return material design questions to the same WorkRef.
-- `rsp-design`: resolve one tracked domain-model, module/seam, or reversible-exploration question from project evidence, writing only authorized planned design back to the selected Change.
-- `rsp-implement`: implement one selected, ready Change within explicit mutation authority and return truthful Tasks, Blockers, and fresh verification evidence.
-- `rsp-diagnose`: establish a confirmed cause or truthful unresolved diagnosis before production correction.
-- `rsp-tdd`: drive one clear behavior through observed RED, minimal GREEN, optional safe REFACTOR, and fresh verification.
-- `rsp-review`: read-only review of code, document, or mixed Changes against fixed scope and project authority.
-- `rsp-address-review`: dispose fixed review findings, apply only authorized accepted corrections, require fresh verification and report-only re-review, and return a recoverable artifact-scoped continuation when work remains.
-- `rsp-release-docs`: prepare or audit evidence-based changelogs, release notes, and migration notes while adapting to user requirements and repository conventions.
-- `rsp-manage`: continue one explicitly requested focused Change across genuinely independent slices or interruption recovery, while keeping ordinary work direct and orchestration state transient.
+Response language and artifact language are independent. Human-facing response headings, labels, explanations, and conclusions follow the requested response language, response-specific project instructions, then the conversation language. Authorized artifact prose follows the requested artifact language, artifact-specific project instructions, then the existing artifact language, and only then the conversation language. Canonical RSP artifact headings, WorkRef values, paths, commands, identifiers, and machine-consumed values remain unchanged; response labels may retain technical tokens in parentheses but never use them as untranslated labels.
 
-Each Skill is independently invocable and returns results to existing project or RSP artifact owners. The suite adds no hidden workflow state or recursive Skill orchestration, and no Skill infers commit, push, or publication authority.
+Compose the suite from evidence: an explicit bounded question may enter report-only Pre-Change Design before Shape; otherwise Shape settles the executable owner and Tracked Design returns one material question to it. Core then chooses Diagnose, TDD, or Implement; Review stays report-only; Address Review corrects accepted findings and requests re-review; Core performs the durable decision before archive. An explicit release operation with a confirmed identity or range may enter Release Docs without a Release Change; create one only for material decisions, coordination, recovery, blockers, or acceptance. Manage is optional: it accepts one selected ready Change or shallow Group that needs independent dispatch, long continuation, or recovery, while small or coupled work stays direct. Projects may keep explicit activation or let Core select eligible managed work automatically.
 
-Response language and artifact language are independent. Human-facing response headings, labels, explanations, and conclusions follow the requested response language, response-specific project instructions, then the conversation language. Authorized artifact prose follows the requested artifact language, artifact-specific project instructions, then the existing artifact language, and only then the conversation language. Canonical RSP artifact headings, WorkRef values, paths, commands, identifiers, and machine-consumed values remain unchanged; response labels may retain technical tokens in parentheses but never use them as untranslated labels.
+### Managed automation policy
+
+Configure automatic selection and local closeout independently in `.rsp/config.yaml`:
+
+```yaml
+manage:
+  activation: auto
+  closeout: lifecycle
+```
+
+`activation` is `explicit` or `auto`. `auto` authorizes Core to select Manage for an eligible requested completion or continuation; it does not grant planning, product-mutation, lifecycle, Git, or external authority. `closeout` is one of:
+
+- `manual`: neither archive nor commit is automatic.
+- `lifecycle`: archive may follow durable review, but commit still needs separate authority.
+- `local`: lifecycle closeout plus the existing bounded local checkpoint or terminal-commit path may run only after its exact-path, clean-boundary, verification, and delivery-value gates.
+
+New `rsp init` config templates use the shown `auto` plus `lifecycle` policy. If `manage` is absent, RSP instead resolves `activation: explicit` and `closeout: local` for compatibility with the previous explicitly requested Manage behavior. Plain `rsp status` and the top-level `manage` object in `rsp status --json` show the resolved values. Nearest scoped restrictions and host enforcement can narrow the configured ceiling. RSP intentionally has no `full` preset: push, tag, publication, deployment, approval, and human acceptance remain explicit and external.
+
+Humans should start with this README. Agents should follow the nearest `AGENTS.md`, prefer `skills/rsp/SKILL.md`, and use `.rsp/rsp-rules.md` only when the Skill is unavailable.
 
-The 3.1 beta product surface contains these ten Skills. The optional `rsp-manage` capability activates only on explicit request and only for a focused ready Change with genuinely independent slices, long continuation, or interruption recovery. Small or coupled work returns directly to Core or the selected Discipline without a dispatch envelope. Managed runs keep authority in current project and RSP artifacts, keep process chronology transient, use bounded verification, and stop at mutation, lifecycle, Git, publication, environment, or human-decision boundaries.
-
-For a complete tracked change, compose the suite as evidence requires: `rsp-shape` returns an executable Change and sends one material design question to `rsp-design`; `rsp-design` returns its evidence, recommendation, alternatives, and any authorized planned-design update to the same WorkRef; Core routes unexplained failures to `rsp-diagnose`, explicit or concrete-risk test-first work to `rsp-tdd`, and ordinary evidenced edits to `rsp-implement`; `rsp-review` returns a read-only report; `rsp-address-review` disposes findings and returns accepted corrections through fresh verification and re-review; when a selected Change explicitly owns a confirmed release identity or range and unfinished release documentation, Core routes that work to `rsp-release-docs`, which turns one evidence ledger into distinct changelog, release-note, and migration surfaces without inferring publication authority; and `rsp` routes implemented current facts, lasting rationale, project-owned context or instructions, and temporary continuation to their existing semantic owners before archive. Use `rsp-manage` only when the user explicitly requests managed continuation and the work qualifies for bounded independent dispatch. Each discipline returns to its existing owner. Ambiguity, failed gates, missing authority, and out-of-scope Git conflicts stop at that owner; no Skill infers Git continuation, commit, delivery, or an automatic retry loop.
-
-Reading guidance:
-
-- `README.md`: human-oriented overview and examples
-- `.rsp/rsp-rules.md`: minimal fallback protocol when the skill is unavailable
-- `skills/rsp/SKILL.md`: preferred operational guide for agents
-- `skills/rsp-shape/SKILL.md`: bounded shaping and slicing guidance
-- `skills/rsp-design/SKILL.md`: tracked domain, module/seam, and reversible-exploration design guidance
-- `skills/rsp-implement/SKILL.md`: bounded implementation and fresh verification guidance
-- `skills/rsp-diagnose/SKILL.md`: evidence-backed cause isolation before correction
-- `skills/rsp-tdd/SKILL.md`: bounded red-green-refactor guidance
-- `skills/rsp-review/SKILL.md`: read-only review guidance
-- `skills/rsp-address-review/SKILL.md`: review finding disposition, correction, re-review, and recovery guidance
-- `skills/rsp-release-docs/SKILL.md`: evidence-based changelog, release-note, and migration guidance
-- `skills/rsp-manage/SKILL.md`: explicit bounded continuation across independent slices
-
-Surface matrix:
-
-| Surface | Primary audience | Role |
-|---|---|---|
-| `README.md` | Humans | Overview, onboarding, examples |
-| `.rsp/rsp-rules.md` | Agents without the skill | Minimal tool-agnostic fallback protocol |
-| `skills/rsp/SKILL.md` | Agents | Preferred operational guide |
-| `skills/rsp-shape/SKILL.md` | Agents | Shape one executable Change or justified shallow Group |
-| `skills/rsp-design/SKILL.md` | Agents | Resolve one tracked design question and return it to the same WorkRef |
-| `skills/rsp-implement/SKILL.md` | Agents | Implement one ready Change with fresh verification evidence |
-| `skills/rsp-diagnose/SKILL.md` | Agents | Confirm a cause before production correction |
-| `skills/rsp-tdd/SKILL.md` | Agents | Implement one clear behavior test-first |
-| `skills/rsp-review/SKILL.md` | Agents | Read-only Code and Document review |
-| `skills/rsp-address-review/SKILL.md` | Agents | Resolve review findings and return a recoverable continuation |
-| `skills/rsp-release-docs/SKILL.md` | Agents | Prepare or audit convention-compatible release documentation |
-| `skills/rsp-manage/SKILL.md` | Agents | Continue eligible independent slices only after explicit invocation |
-| `AGENTS.md` | Humans and agents | Scoped project instructions and RSP navigation |
-
-Humans should usually start with `README.md`; agents should follow nearest `AGENTS.md`, load the `rsp` skill when available, and use `.rsp/rsp-rules.md` only as fallback.
-
-When this README shows `rsp <command>`, it assumes the command is already available in your environment. For this opt-in beta, otherwise use the exact `npx -y @oevery/rsp@3.1.0-beta.0 <command>` identity after publication; stable users can keep using the unversioned package entrypoint for npm `latest`.
-
-Install the exact package-bundled suite into the current project:
+When this README shows `rsp <command>`, it assumes the command is already available in your environment. For opt-in beta evaluation, pin the exact prerelease identity, such as `npx -y @oevery/rsp@3.1.0-beta.2 <command>`; stable users can keep using the unversioned package entrypoint for npm `latest`.
+
+Install the exact package's default lifecycle suite into the current project:
 
 ```bash
 rsp skills install --dry-run
 rsp skills install
 ```
 
-The command preflights all ten package-owned targets, leaves unrelated `.agents/skills` entries untouched, and requires explicit `--force` before replacing a divergent package-owned directory. It installs from the package that invoked `rsp`, so prerelease dogfooding can use one exact npm identity after publication:
+On a dual TTY, `rsp skills` opens the interactive manager: the default suite is selected and locked, while optional project Skills can be selected before one confirmed atomic install. For scripts, redirected terminals, and CI, use the deterministic discovery and install commands:
+
+```bash
+rsp skills list
+rsp skills list --json
+```
+
+The command preflights the eleven default package-owned targets, leaves unrelated `.agents/skills` entries (including optional Skills) untouched, and requires explicit `--force` before replacing a divergent selected directory. Install the optional project-level audit Skill by exact name:
+
+```bash
+rsp skills install rsp-codebase-audit --dry-run
+rsp skills install rsp-codebase-audit
+```
+
+Both forms install from the package that invoked `rsp`, so prerelease dogfooding can pin one exact npm identity:
 
 ```bash
-npx -y @oevery/rsp@3.1.0-beta.0 skills install --dry-run
-npx -y @oevery/rsp@3.1.0-beta.0 skills install
+npx -y @oevery/rsp@3.1.0-beta.2 skills install --dry-run
+npx -y @oevery/rsp@3.1.0-beta.2 skills install
 ```
 
-These registry commands remain unavailable until `3.1.0-beta.0` is published. `rsp update` refreshes RSP-managed project files only; run `rsp skills install` separately to refresh the package-owned Skill suite.
+Exact prerelease identities avoid depending on a moving dist-tag. `rsp update` refreshes RSP-managed project files only; run `rsp skills install` separately to refresh the package-owned Skill suite.
 
 ## Migrating from 2.x
 
@@ -272,8 +273,8 @@ Durable review makes two independent semantic choices: whether current facts or
 
 New project:
 
-1. `npx -y @oevery/rsp@3.1.0-beta.0 init`
-2. Prefer `npx -y @oevery/rsp@3.1.0-beta.0 init --with-project-setup`, or run `rsp create project-setup` manually
+1. `npx -y @oevery/rsp@3.1.0-beta.2 init`
+2. Prefer `npx -y @oevery/rsp@3.1.0-beta.2 init --with-project-setup`, or run `rsp create project-setup` manually
 3. Fill `.rsp/specs/design.md`
 4. Use `rsp add spec <name>` only when a new durable project doc is needed
 5. Keep lasting rationale in the configured Decision Record directory and stable scoped operating instructions in the nearest project-owned `AGENTS.md`
@@ -286,14 +287,14 @@ New project:
 
 Existing project with a rich `AGENTS.md`:
 
-1. `npx -y @oevery/rsp@3.1.0-beta.0 init`
+1. `npx -y @oevery/rsp@3.1.0-beta.2 init`
 2. Keep the managed block thin
-3. Move durable design into `.rsp/specs/design.md`
+3. Keep project-wide boundaries and navigation in `.rsp/specs/design.md`, and route cohesive durable facts to the smallest domain Spec
 4. Use `rsp add spec <name>` only when another durable current-fact document is needed
 
 AI-assisted setup:
 
-1. `npx -y @oevery/rsp@3.1.0-beta.0 init --agents-mode print --with-project-setup`
+1. `npx -y @oevery/rsp@3.1.0-beta.2 init --agents-mode print --with-project-setup`
 2. Keep the managed block as generated and adapt only the surrounding human-owned content if needed
 3. Have the AI review and fill `.rsp/changes/project-setup.md`
 4. Have the AI fill `.rsp/specs/design.md`
@@ -306,8 +307,10 @@ rsp init --agents-mode <mode>   Scaffold .rsp/ and ensure AGENTS.md contains the
 rsp init --with-project-setup   Also create .rsp/changes/project-setup.md
 rsp update                      Refresh the fallback protocol, repair the AGENTS block, and rebuild indices
 rsp ui [--lang auto|en|zh-CN]   Open the read-only interactive dashboard
-rsp skills install [--dry-run] [--force]
-                                  Install the invoking package's ten Skills into .agents/skills
+rsp skills                      Open the interactive project Skill manager on a dual TTY
+rsp skills list [--json]        List bundled Skills and exact project installation status
+rsp skills install [name] [--dry-run] [--force]
+                                  Install the default eleven Skills or one exact optional Skill into .agents/skills
 rsp add spec <name>             Create .rsp/specs/<name>.md and rebuild specs index
 rsp create <name> [summary]     Create .rsp/changes/<name>.md; add --lite for a shorter template
 rsp group create <name> [goal] Create an unfocused .rsp/changes/<name>/00-brief.md

package/README.zh-CN.md

@@ -4,7 +4,7 @@
 
 **面向人类与 AI Agent 的仓库原生工程工作流。**
 
-RSP 将模糊意图逐步转化为边界明确、经过实现、审查和验证的软件变更，同时让项目知识可以长期维护、让未完成工作可以可靠恢复。十个可组合 Skills 共同覆盖工程旅程，但不引入隐藏 workflow state，也不取代项目自己的文件、工具和权限边界。
+RSP 将模糊意图逐步转化为边界明确、经过实现、审查和验证的软件变更，同时让项目知识可以长期维护、让未完成工作可以可靠恢复。十一个可组合 Skills 共同覆盖工程旅程，但不引入隐藏 workflow state，也不取代项目自己的文件、工具和权限边界。
 
 这套工作流建立在轻量的 **Rules、Specs、Plans** artifact foundation 之上。
 
@@ -15,7 +15,7 @@ RSP 将模糊意图逐步转化为边界明确、经过实现、审查和验证
 - 将原因不明的失败交给 diagnosis，将普通且证据充分的修改交给 implementation，仅将明确要求或具有具体风险的 test-first 工作交给 TDD。
 - 按明确范围审查变更、处理已接受的 findings，并重新运行 fresh verification。
 - 将稳定事实、作用域指令、长期理由和完成历史分别保存在正确的 owner 中。
-- 当 Change 明确拥有已确认的 release 时，准备对应的发布文档。
+- 为显式且已确认的 release operation 准备或校准发布文档；只有重大协调需要持久 owner 时才使用 Release Change。
 
 ## 工作流如何组合
 
@@ -25,7 +25,7 @@ RSP 将模糊意图逐步转化为边界明确、经过实现、审查和验证
   → 按需 Design
   → Diagnose | TDD | Implement
   → Review → Address accepted findings
-  → 按需准备 Release Docs
+  → 显式且确认后的 Release Docs
   → Durable Review → Archive
 ```
 
@@ -34,17 +34,17 @@ RSP 根据 selected Change、仓库证据、验证结果和 blockers 推导下
 ## 快速开始
 
 ```bash
-npx -y @oevery/rsp@3.1.0-beta.0 init
-npx -y @oevery/rsp@3.1.0-beta.0 doctor
+npx -y @oevery/rsp@3.1.0-beta.2 init
+npx -y @oevery/rsp@3.1.0-beta.2 doctor
 ```
 
 推荐启动流程：
 
 ```bash
-npx -y @oevery/rsp@3.1.0-beta.0 init --with-project-setup
+npx -y @oevery/rsp@3.1.0-beta.2 init --with-project-setup
 # 填写 .rsp/changes/project-setup.md
 # 填写 .rsp/specs/design.md
-npx -y @oevery/rsp@3.1.0-beta.0 doctor
+npx -y @oevery/rsp@3.1.0-beta.2 doctor
 ```
 
 ## Artifact foundation
@@ -105,7 +105,7 @@ npx -y @oevery/rsp@3.1.0-beta.0 doctor
 - `.rsp/specs/design.md`：由 `rsp init` 创建，之后由项目维护。
 - `.rsp/specs/decisions/`：默认的权威 Decision Record 目录；只有 Host Project 已有外部 ADR 目录时才配置 `decisions.path`。
 - `.rsp/rsp-rules.md`：生成的最小 fallback protocol；可用时优先加载 `rsp` skill。
-- 将长期架构、边界和跨模块技术约束放在 `.rsp/specs/design.md`。
+- 将项目级边界与导航放在 `.rsp/specs/design.md`；把成组且可复用的事实放入 `.rsp/specs/INDEX.md` 所列的最小领域 Spec。
 - 将 `.rsp/specs/INDEX.md` 视为附加 spec 的目录；它不列出 `design.md`。
 - 将稳定且有作用域的工作流、验证和本地运行指令放在 nearest project-owned `AGENTS.md` 的非受管区域。
 
@@ -149,64 +149,80 @@ Do not treat `.rsp/specs/` or `.rsp/changes/` as replacements for nearest `AGENT
 
 ## Skills
 
-RSP 发布十个宿主无关、按需加载的 Skills：
+RSP 发布十一个默认、宿主无关、按需加载的生命周期 Skills，以及一个需要显式安装的项目级审计 Skill：
+
+| Skill | 职责 |
+|---|---|
+| `rsp` | 派生下一动作，指导 setup、durable review 与 archive 判断。 |
+| `rsp-shape` | 在不实现的前提下塑造一个可执行 Change 或合理的 shallow Group。 |
+| `rsp-design` | 解决一个受跟踪的 domain、module/seam 或 reversible-exploration 问题。 |
+| `rsp-implement` | 实现一个 selected ready Change，并提供 fresh verification。 |
+| `rsp-diagnose` | 修正前确认原因，或如实返回 unresolved diagnosis。 |
+| `rsp-tdd` | 让一个清晰行为经过 RED、GREEN 与安全的 REFACTOR。 |
+| `rsp-review` | 对固定的 Code、Document 或 mixed comparison 做只读审查。 |
+| `rsp-address-review` | 处置固定 findings，修正 accepted 项，验证并请求复审。 |
+| `rsp-commit` | 使用与仓库一致的结构化消息创建一个已授权、精确范围的本地提交。 |
+| `rsp-release-docs` | 起草、审计、定稿或校准基于证据的发布表面。 |
+| `rsp-manage` | 继续一个显式请求或项目启用且符合条件的 ready Change 或 shallow Group。 |
+| `rsp-codebase-audit`（可选） | 只读审计一个有界仓库或子树，发现有证据支撑的结构风险。 |
+
+每个生命周期 Skill 都把结果返回现有项目或 RSP owner。只读的 Pre-Change Design 与 `rsp-codebase-audit` 可以把一个有界结果直接返回用户，而不虚构 artifact owner。套件不引入隐藏 workflow state 或递归 Skill 编排。任何 Skill 都不会推断 commit、push、publication、deployment、approval 或 human-acceptance 权限。
 
-- `rsp`：setup、workflow、durable review 和 archive 指导。
-- `rsp-shape`：在不实现代码的前提下，将不清晰的非平凡工作塑造成一个 ready Change 或合理的 shallow Group；当用户明确要求严格质询或高风险决策仍未解决时，渐进加载 deep clarification，并让关键设计问题通过同一个 WorkRef 返回。
-- `rsp-design`：根据项目证据解决一个受跟踪的 domain model、module/seam 或 reversible exploration 问题，只把已授权的 planned design 写回 selected Change。
-- `rsp-implement`：在显式 mutation authority 内实现一个 selected、ready Change，并如实回写 Tasks、Blockers 和 fresh verification evidence。
-- `rsp-diagnose`：在生产修复前确认原因，或如实返回 unresolved diagnosis。
-- `rsp-tdd`：让一个清晰行为经过 observed RED、minimal GREEN、可选 safe REFACTOR 与 fresh verification。
-- `rsp-review`：基于固定范围与项目权威，对 Code、Document 或 mixed Change 进行只读审查。
-- `rsp-address-review`：处置固定 review findings，仅修复已授权且 accepted 的 finding，并要求 fresh verification 与 report-only re-review。
-- `rsp-release-docs`：根据证据准备或审计 Changelog、Release Notes 和 Migration Notes，并适配用户要求与仓库现有约定。
-- `rsp-manage`：仅在用户显式请求后，跨真正独立的 slices 或中断恢复继续一个 focused Change；普通工作保持直接执行，编排状态保持临时。
+响应语言与产物语言相互独立。面向人的响应标题、标签、解释和结论依次遵循明确指定的响应语言、项目中针对响应的指令和会话语言；已授权写入的产物正文依次遵循明确指定的产物语言、项目中针对产物的指令、目标产物的现有语言，最后才回退到会话语言。RSP 的 canonical artifact headings、WorkRef 值、路径、命令、标识符和机器消费值保持不变；响应标签可以在括号中保留技术 token，但不能直接使用未翻译的 token 作为标签。
 
-每个 Skill 都可以独立调用，并把结果返回现有项目或 RSP artifact owner。套件不引入隐藏 workflow state 或递归 Skill 编排，也不会由任何 Skill 推断 commit、push 或 publication 权限。
+按证据组合套件：Shape 确定 owner；Design 返回一个关键问题；Core 在 Diagnose、TDD 与 Implement 之间选择；Review 保持只读；Address Review 修正 accepted findings 并请求复审；Core 在 archive 前执行 durable decision。显式 release operation 在 identity 或 range 已确认时，可以不创建 Release Change 就进入 Release Docs；只有重大决策、协调、恢复、blocker 或 acceptance 需要持久 owner 时才创建。Manage 是可选能力：它接受一个 selected ready Change 或 shallow Group，并要求独立调度、长时 continuation 或恢复需求；小型或紧耦合工作保持直接执行。项目可以保持显式启用，也可以让 Core 自动选择符合条件的 managed work。
 
-响应语言与产物语言相互独立。面向人的响应标题、标签、解释和结论依次遵循明确指定的响应语言、项目中针对响应的指令和会话语言；已授权写入的产物正文依次遵循明确指定的产物语言、项目中针对产物的指令、目标产物的现有语言，最后才回退到会话语言。RSP 的 canonical artifact headings、WorkRef 值、路径、命令、标识符和机器消费值保持不变；响应标签可以在括号中保留技术 token，但不能直接使用未翻译的 token 作为标签。
+### Managed automation policy
 
-3.1 beta 产品面包含这十个 Skills。可选的 `rsp-manage` 仅在显式请求后启用，并要求 focused ready Change 具有真正独立的 slices、长时 continuation 或中断恢复需求。小型或紧耦合工作直接返回 Core 或对应 Discipline，不产生 dispatch envelope。managed run 以当前项目和 RSP artifacts 为 authority，过程 chronology 不持久化，验证保持有界，并在 mutation、lifecycle、Git、publication、environment 或 human-decision authority 边界停止。
+在 `.rsp/config.yaml` 中分别配置自动选择与本地收尾：
 
-完成一个 tracked Change 时，应按证据组合套件：`rsp-shape` 返回可执行 Change，并把一个关键设计问题交给 `rsp-design`；`rsp-design` 把证据、建议、备选方案和已授权的 planned-design 更新返回同一个 WorkRef；Core 把 unexplained failure 路由到 `rsp-diagnose`，仅把明确要求或具有具体风险的 test-first 工作路由到 `rsp-tdd`，把普通且证据充分的修改路由到 `rsp-implement`；`rsp-review` 返回只读报告；`rsp-address-review` 处置 finding；当 selected Change 明确拥有已确认的 release identity 或 range，且发布文档尚未完成时，Core 才把工作路由到 `rsp-release-docs`；最后由 `rsp` 在 archive 前把 stable facts、lasting rationale 与 temporary continuation 路由到各自 owner。只有用户显式要求 managed continuation 且工作满足有界独立调度条件时才使用 `rsp-manage`。遇到歧义、失败门禁、缺失权限或越界 Git conflict 时停止；任何 Skill 都不会推断 Git continuation、commit、delivery 权限或自动重试。
+```yaml
+manage:
+  activation: auto
+  closeout: lifecycle
+```
 
-文档分层矩阵：
+`activation` 可取 `explicit` 或 `auto`。`auto` 只授权 Core 为符合条件且已经请求完成或继续的工作选择 Manage；它不授予 planning、product mutation、lifecycle、Git 或外部权限。`closeout` 可取：
 
-| Surface | 主要受众 | 职责 |
-|---|---|---|
-| `README.md` | 人类 | 概览、入门、示例 |
-| `.rsp/rsp-rules.md` | 未安装 skill 的 agent | 最小 tool-agnostic fallback protocol |
-| `skills/rsp/SKILL.md` | agent | 首选操作指南 |
-| `skills/rsp-shape/SKILL.md` | agent | 塑造一个可执行 Change 或合理的 shallow Group |
-| `skills/rsp-design/SKILL.md` | agent | 解决一个受跟踪的设计问题并返回同一个 WorkRef |
-| `skills/rsp-implement/SKILL.md` | agent | 用 fresh verification evidence 实现一个 ready Change |
-| `skills/rsp-diagnose/SKILL.md` | agent | 在生产修复前确认原因 |
-| `skills/rsp-tdd/SKILL.md` | agent | 以 test-first 方式实现一个清晰行为 |
-| `skills/rsp-review/SKILL.md` | agent | Code 与 Document 只读审查 |
-| `skills/rsp-address-review/SKILL.md` | agent | 处置 review findings 并返回可恢复 continuation |
-| `skills/rsp-release-docs/SKILL.md` | agent | 准备或审计符合项目约定的发布文档 |
-| `skills/rsp-manage/SKILL.md` | agent | 仅在显式调用后继续符合条件的独立 slices |
-| `AGENTS.md` | 人类与 agent | 有作用域的项目指令与 RSP 导航入口 |
+- `manual`：archive 与 commit 都不会自动执行。
+- `lifecycle`：durable review 后可以 archive，但 commit 仍需要独立权限。
+- `local`：允许 lifecycle closeout，并仅在 exact-path、clean-boundary、verification 和 delivery-value gate 通过后进入已有的有界本地 checkpoint 或 terminal commit 路径。
 
-通常应由人先读 `README.md`；agent 应遵循 nearest `AGENTS.md`，可用时加载 `rsp` skill，仅在 skill 不可用时读取 `.rsp/rsp-rules.md`。
+新的 `rsp init` config template 使用上面的 `auto` 加 `lifecycle` policy。如果省略 `manage`，RSP 则为兼容既有显式 Manage 行为解析为 `activation: explicit` 与 `closeout: local`。普通 `rsp status` 和 `rsp status --json` 顶层的 `manage` 对象都会显示解析后的值。nearest scoped restriction 与 host enforcement 可以缩小该配置上限。RSP 有意不提供 `full` preset：push、tag、publication、deployment、approval 与 human acceptance 始终保持显式且属于外部边界。
 
-如果文档中写的是 `rsp <command>`，默认前提是你的环境里已经能直接运行 `rsp`。对本次 opt-in beta，发布后应使用精确身份 `npx -y @oevery/rsp@3.1.0-beta.0 <command>`；stable 用户仍可使用不带版本的入口获取 npm `latest`。
+人类应从本 README 开始；agent 应遵循 nearest `AGENTS.md`，优先加载 `skills/rsp/SKILL.md`，仅在 Skill 不可用时读取 `.rsp/rsp-rules.md`。
 
-将当前精确包内置的完整 Skill 套件安装到当前项目：
+如果文档中写的是 `rsp <command>`，默认前提是你的环境里已经能直接运行 `rsp`。进行 opt-in beta 评估时，应固定精确 prerelease 身份，例如 `npx -y @oevery/rsp@3.1.0-beta.2 <command>`；stable 用户仍可使用不带版本的入口获取 npm `latest`。
+
+将当前精确包内置的默认生命周期 Skill 套件安装到当前项目：
 
 ```bash
 rsp skills install --dry-run
 rsp skills install
 ```
 
-该命令会预检全部十个 package-owned targets，保留无关的 `.agents/skills` 条目，并且只有显式传入 `--force` 才会替换内容不同的 package-owned 目录。它从调用 `rsp` 的同一个包安装，因此 beta 发布后可以固定精确 npm 身份：
+在 dual TTY 中，`rsp skills` 会打开交互式管理器：默认套件保持选中并锁定，可在确认一次原子安装前选择可选的项目级 Skill。脚本、重定向终端和 CI 使用确定性的发现与安装命令：
+
+```bash
+rsp skills list
+rsp skills list --json
+```
+
+该命令会预检十一个默认 package-owned targets，保留无关的 `.agents/skills` 条目（包括可选 Skills），并且只有显式传入 `--force` 才会替换内容不同的已选目录。按精确名称安装可选的项目级审计 Skill：
+
+```bash
+rsp skills install rsp-codebase-audit --dry-run
+rsp skills install rsp-codebase-audit
+```
+
+两种形式都从调用 `rsp` 的同一个包安装，因此 beta 评估可以固定精确 npm 身份：
 
 ```bash
-npx -y @oevery/rsp@3.1.0-beta.0 skills install --dry-run
-npx -y @oevery/rsp@3.1.0-beta.0 skills install
+npx -y @oevery/rsp@3.1.0-beta.2 skills install --dry-run
+npx -y @oevery/rsp@3.1.0-beta.2 skills install
 ```
 
-在 `3.1.0-beta.0` 实际发布前，上述 registry 命令不可用。`rsp update` 只刷新由 RSP 管理的项目文件；刷新 package-owned Skill 套件需要单独运行 `rsp skills install`。
+精确 prerelease 身份可以避免依赖持续移动的 dist-tag。`rsp update` 只刷新由 RSP 管理的项目文件；刷新 package-owned Skill 套件需要单独运行 `rsp skills install`。
 
 ## 从 2.x 迁移
 
@@ -257,8 +273,8 @@ durable review 包含两个独立语义判断：是否更新当前事实或作
 
 新项目：
 
-1. `npx -y @oevery/rsp@3.1.0-beta.0 init`
-2. 优先使用 `npx -y @oevery/rsp@3.1.0-beta.0 init --with-project-setup`，或手动执行 `rsp create project-setup`
+1. `npx -y @oevery/rsp@3.1.0-beta.2 init`
+2. 优先使用 `npx -y @oevery/rsp@3.1.0-beta.2 init --with-project-setup`，或手动执行 `rsp create project-setup`
 3. 填写 `.rsp/specs/design.md`
 4. 仅在需要新的长期项目文档时使用 `rsp add spec <name>`
 5. 将长期理由写入配置的 Decision Record 目录，将稳定的作用域操作指令写入 nearest project-owned `AGENTS.md`
@@ -271,14 +287,14 @@ durable review 包含两个独立语义判断：是否更新当前事实或作
 
 已有复杂 `AGENTS.md` 的项目：
 
-1. `npx -y @oevery/rsp@3.1.0-beta.0 init`
+1. `npx -y @oevery/rsp@3.1.0-beta.2 init`
 2. 保持受管块尽量薄
-3. 将长期设计收敛到 `.rsp/specs/design.md`
+3. 将项目级边界与导航放在 `.rsp/specs/design.md`，把成组的长期事实放入最小领域 Spec
 4. 只有需要额外的 durable current-fact 文档时才使用 `rsp add spec <name>`
 
 AI 协助接入：
 
-1. `npx -y @oevery/rsp@3.1.0-beta.0 init --agents-mode print --with-project-setup`
+1. `npx -y @oevery/rsp@3.1.0-beta.2 init --agents-mode print --with-project-setup`
 2. 保持受管块原样，只在需要时调整周围由人维护的内容
 3. 让 AI 审阅并填写 `.rsp/changes/project-setup.md`
 4. 让 AI 填写 `.rsp/specs/design.md`
@@ -291,8 +307,10 @@ rsp init --agents-mode <mode>   搭建 .rsp/，并确保 AGENTS.md 含有 RSP
 rsp init --with-project-setup   同时创建 .rsp/changes/project-setup.md
 rsp update                      刷新 fallback protocol、修复 AGENTS 受管块并重建索引
 rsp ui [--lang auto|en|zh-CN]   打开只读交互式仪表盘
-rsp skills install [--dry-run] [--force]
-                                  将当前包的十个 Skills 安装到 .agents/skills
+rsp skills                      在 dual TTY 中打开交互式项目 Skill 管理器
+rsp skills list [--json]        列出内置 Skills 及其精确项目安装状态
+rsp skills install [name] [--dry-run] [--force]
+                                  将默认十一个 Skills 或一个精确可选 Skill 安装到 .agents/skills
 rsp add spec <name>             创建 .rsp/specs/<name>.md 并重建 specs 索引
 rsp create <name> [summary]     创建 .rsp/changes/<name>.md；可加 --lite 使用更短模板
 rsp group create <name> [goal] 创建不进入 focus 的 .rsp/changes/<name>/00-brief.md