cowork-flow 2.3.0 → 2.4.0

package/template/.cowork-flow/spec/references/orchestration-patterns.md

@@ -1,141 +1,141 @@
 # Orchestration Patterns
 
-> 已实现的协作模式声明性目录。
-> 本文件是 spec/contracts/ 系列的人可读摘要——权威定义以 spec/contracts/ 为准。
+> Declarative catalog of the collaboration patterns already implemented.
+> This file is a human-readable summary of the `spec/contracts/` series — the contracts are authoritative.
 
-## 核心原则
+## Core principle
 
-**主会话是唯一的协调者。固定子代理是叶子执行者。用户是流程的最终决策者。**
+**The main session is the only coordinator. Fixed subagents are leaf executors. The user is the final decision-maker.**
 
 ---
 
-## 模式 1：叶子执行器（已实现）
+## Pattern 1: Leaf executors (implemented)
 
 ```
-主会话 -> cowork-implement（实现）-> 返回 -> 主会话验证
-主会话 -> cowork-check（验证）     -> 返回 -> 主会话确认
-主会话 -> cowork-research（调研）  -> 返回 -> 主会话综合
+main session -> cowork-implement (implement) -> return -> main session verifies
+main session -> cowork-check (verify)        -> return -> main session confirms
+main session -> cowork-research (research)   -> return -> main session synthesizes
 ```
 
-**使用场景**: 计划声明了并行低冲突切片（文件所有权干净、各有独立验证）；需要上下文隔离的调研（cowork-research）；完成前必须的独立复核（cowork-check）。
+**When to use**: the plan declares parallel low-conflict slices (clean file ownership, independent verification); research that needs context isolation (cowork-research); the independent review required before completion (cowork-check).
 
-**约束**:
-- R-AG-001: 子代理不能派发其他子代理
-- R-AG-002: 子代理不修改 spec 文件
-- R-AG-003: 子代理不操作 Git
-- R-AG-004: 子代理不提交/归档
+**Constraints**:
+- R-AG-001: subagents must not dispatch other subagents
+- R-AG-002: subagents must not modify spec files
+- R-AG-003: subagents must not operate Git
+- R-AG-004: subagents must not commit/archive
 
-**验证**: dispatch 前用 `subagent init --role` 创建 runtime context；child 第一个动作必须是 `subagent bind`；parent 检查 `status: bound`。
+**Verification**: create the runtime context with `subagent init --role` before dispatch; the child's first action must be `subagent bind`; the parent checks `status: bound`.
 
 ---
 
-## 模式 2：并行扇出 + 合并（已实现，主会话管理）
+## Pattern 2: Parallel fan-out + merge (implemented, main-session managed)
 
 ```
-主会话 ─┬─> cowork-implement (task-slice-a) ─┐
-        ├─> cowork-implement (task-slice-b) ─┤─> 所有返回 ─> 集成验证 ─> 主会话综合
-        └─> cowork-research (background)    ─┘
+main session ─┬─> cowork-implement (task-slice-a) ─┐
+              ├─> cowork-implement (task-slice-b) ─┤─> all return ─> integrated verification ─> main session synthesizes
+              └─> cowork-research (background)    ─┘
 ```
 
-**使用场景**: 多个低耦合的实现切片可以并行。
+**When to use**: multiple low-coupling implementation slices can run in parallel.
 
-**验证清单**:
-1. 子任务之间无写冲突（不同文件或通过 git worktree 隔离）
-2. 每个子代理产出不同类型的发现
-3. 合并步骤在主 agent 的上下文窗口内可完成
-4. 用户等待时长足够使 parallel wall-clock 收益明显
-5. 收束为单次 cowork-check 复核与单次提交（回执指纹覆盖合并后的变更集）
-6. 同一工作树不能并行两个任务（文件范围门禁读取工作树全量变更）；跨任务并行使用独立 worktree/clone 并经看板依赖边协调
+**Verification checklist**:
+1. No write conflicts between subtasks (different files or isolated through git worktree)
+2. Each subagent produces a different type of finding
+3. The merge step fits within the main agent's context window
+4. The user's wait is long enough for parallel wall-clock gains to matter
+5. Converge on a single cowork-check review and a single commit (the receipt fingerprint covers the merged change set)
+6. Two tasks cannot run in parallel in one worktree (the file-scope gate reads the whole working tree); cross-task parallelism uses separate worktrees/clones and coordinates through board dependency edges
 
-任一答案为否 → 回退为模式 1 顺序执行。
+If any answer is no, fall back to Pattern 1 sequential execution.
 
 ---
 
-## 模式 3：顺序 skill 链（已实现，用户驱动）
+## Pattern 3: Sequential skill chain (implemented, user-driven)
 
 ```
-用户执行: task next --run --title ... -> brainstorming -> task-planning -> task next <dir> --run -> implement -> check -> task next <dir> --run --intent review
+User runs: task next --run --title ... -> brainstorming -> task-planning -> task next <dir> --run -> implement -> check -> task next <dir> --run --intent review
 ```
 
-**使用场景**: 阶段之间有依赖关系，用户在阶段间判断有价值。
+**When to use**: stages depend on each other, and user judgment between stages is valuable.
 
-**为什么不做生命周期编排 agent**:
-- 编排 agent 会在 hand-off 间总结→丢失细微差异
-- 用户在阶段间的检查点能早期发现方向错误
-- 人工检查的成本 < 走错方向的返工成本
+**Why there is no lifecycle-orchestration agent**:
+- An orchestration agent summarizes between hand-offs, losing nuance
+- User checkpoints between stages catch wrong directions early
+- The cost of a human check is lower than the rework cost of going the wrong way
 
 ---
 
-## 模式 4：Party Mode V2 Board（已实现，咨询性质）
+## Pattern 4: Party Mode V2 Board (implemented, advisory)
 
 ```
-主会话 ─> 启动 Discussion（party-v2 runtime board）
-       ─> 子代理通过 board 交流（不直接对话）
-       ─> 主持人监控漂移，不参与决策
-       ─> 产出建议，不改变任务状态
+main session ─> start Discussion (party-v2 runtime board)
+              ─> subagents communicate through the board (never directly)
+              ─> moderator monitors drift and does not decide
+              ─> produces advice and does not change task state
 ```
 
-**约束**:
-- Board 不能推进任务状态
-- 不能替代 Implement/Check formal gates
-- 主持人不能转述、总结、合成子代理意见
+**Constraints**:
+- The board cannot advance task state
+- It cannot replace the Implement/Check formal gates
+- The moderator must not relay, summarize, or synthesize subagent opinions
 
 ---
 
-## 反模式 A：路由人格（禁止）
+## Anti-pattern A: Router persona (forbidden)
 
 ```
-/workflow -> 代理人格("需要什么角色?") -> 子人格 -> 代理人格(重述) -> 用户
+/workflow -> agent persona ("what role do you need?") -> sub-persona -> agent persona (restates) -> user
 ```
 
-**为什么不行**: 纯路由层没有领域价值; 两次重述→信息丢失+~2x token; AGENTS.md 的 workflow-state 路由已覆盖
+**Why not**: a pure routing layer has no domain value; double restatement loses information and costs ~2x tokens; the `<workflow-state>` routing in AGENTS.md already covers this
 
-**替代方案**: 用 `<workflow-state>` 块直接路由，不通过人格。
+**Alternative**: route directly with the `<workflow-state>` block; do not use personas.
 
 ---
 
-## 反模式 B：人格调用人格（禁止）
+## Anti-pattern B: Persona calling persona (forbidden)
 
 ```
-cowork-check("看到安全相关代码") -> cowork-security-audit
+cowork-check("sees security-related code") -> cowork-security-audit
 ```
 
-**为什么不行**: 人格设计为产出单一视角; 链式调用破坏此设计; 调用人格传递的摘要丢失上下文; 失败模式相乘
+**Why not**: personas are designed to produce a single perspective; chaining breaks that design; summaries passed between personas lose context; failure modes multiply
 
-**替代方案**: 当前人格在报告中推荐后续审计；用户或主会话运行第二轮。
+**Alternative**: the current persona recommends a follow-up audit in its report; the user or main session runs the second round.
 
 ---
 
-## 反模式 C：Deep nesting（禁止）
+## Anti-pattern C: Deep nesting (forbidden)
 
 ```
 /ship -> quality-coordinator -> code-reviewer -> security-auditor
 ```
 
-**为什么不行**: 每一层增加延迟和 token 但无决策价值; 深层人格的上下文多次总结后失真
+**Why not**: each layer adds latency and tokens without decision value; deep persona context degrades after repeated summarization
 
-**替代方案**: cowork-flow 当前深度 = 1（主会话 → 叶子子代理）。不引入中间层。
+**Alternative**: cowork-flow's current depth = 1 (main session -> leaf subagent). Do not introduce intermediate layers.
 
 ---
 
-## 决策流
+## Decision flow
 
 ```
-任务是一个视角对一个 artifact？
-├── 是 → 直接调用。停止。
-└── 否 → task 可并列？
-         ├── 否 → 顺序 skill 链（用户驱动）
-         └── 是 → 并行扇-out（主会话协调）
-                  验证清单：
-                  - 子任务无写冲突
-                  - 每 persona 产出不同类型
-                  - 合并步骤 fit in main context
-                  任一不满足 → 回退顺序
+Is the task one perspective on one artifact?
+├── Yes -> call it directly. Stop.
+└── No  -> can the tasks run in parallel?
+          ├── No  -> sequential skill chain (user-driven)
+          └── Yes -> parallel fan-out (main-session coordinated)
+                     Verification checklist:
+                     - subtasks have no write conflicts
+                     - each persona produces a different type
+                     - the merge step fits in the main context
+                     Any failure -> fall back to sequential
 ```
 
-## 参见
+## See also
 
-- `spec/contracts/subagent-dispatch.md` — formal dispatch 协议权威定义
-- `spec/contracts/party-mode-v2-board.md` — Party Mode V2 权威定义
-- `spec/backend/`、`spec/frontend/`、`spec/guides/` — 用户定义规范，由 review skill 按需逐条审查
-- `skills/cowork-flow/SKILL.md` — workflow state 路由
+- `spec/contracts/subagent-dispatch.md` — authoritative formal dispatch protocol
+- `spec/contracts/party-mode-v2-board.md` — authoritative Party Mode V2 definition
+- `spec/backend/`, `spec/frontend/`, `spec/guides/` — user-defined specs, read clause by clause by the review skill
+- `skills/cowork-flow/SKILL.md` — workflow state routing

package/template/.cowork-flow/spec/references/security-checklist.md

@@ -1,51 +1,51 @@
 # Security Quick Reference
 
-> 安全快速参考。当处理用户输入/认证/数据存储/外部集成时按需加载。
+> Security quick reference. Load on demand when handling user input, authentication, data storage, or external integrations.
 
-契约: ERROR_OUTPUT_AS_DATA_V1 (详见 .cowork-flow/spec/contracts/error-output-as-data.md)
+Contract: ERROR_OUTPUT_AS_DATA_V1 (see .cowork-flow/spec/contracts/error-output-as-data.md)
 
-## 输入验证
+## Input validation
 
-- 所有外部输入在边界验证（API 入口、文件上传、CLI 参数）
-- 错误信息不泄漏内部实现细节（栈追踪、路径、库版本）
-- 参数化查询，禁止字符串拼接 SQL
-- 长度/类型/范围白名单，而非仅黑名单
+- Validate all external input at the boundary (API entry points, file uploads, CLI arguments)
+- Error messages must not leak internal implementation details (stack traces, paths, library versions)
+- Use parameterized queries; never concatenate SQL strings
+- Prefer length/type/range allowlists over blocklists
 
-## 认证授权
+## Authentication and authorization
 
-- 认证状态不在客户端存储（仅 httpOnly secure cookie）
-- 权限检查在每次受保护操作时执行（不依赖前端隐藏）
-- 认证失败不区分用户名错/密码错（统一提示"认证失败"）
-- Session 有过期时间 + rotation on privilege change
+- Authentication state is not stored on the client (httpOnly secure cookie only)
+- Perform permission checks on every protected operation (do not rely on the frontend to hide controls)
+- Authentication failures do not distinguish wrong username from wrong password (use a uniform "authentication failed" message)
+- Sessions have an expiry time and rotate on privilege change
 
-## 依赖与供应链
+## Dependencies and supply chain
 
-- 第三方依赖来自可信来源
-- 无已知 CVE（`pip audit` / `npm audit`）
-- Lockfile 提交到版本控制
+- Third-party dependencies come from trusted sources
+- No known CVEs (`pip audit` / `npm audit`)
+- Lockfiles are committed to version control
 
-## 数据与隐私
+## Data and privacy
 
-- 日志中不出现敏感数据（密码、token、PII）
-- 传输加密（TLS），存储加密（bcrypt/argon2）
-- 错误输出视为不可信数据（参见 ERROR_OUTPUT_AS_DATA_V1）
+- Logs contain no sensitive data (passwords, tokens, PII)
+- Encrypt in transit (TLS) and at rest (bcrypt/argon2)
+- Treat error output as untrusted data (see ERROR_OUTPUT_AS_DATA_V1)
 
-## OWASP Top 10 快速映射
+## OWASP Top 10 quick mapping
 
-| 编号 | 类别 | 检查 | 缺失风险 |
+| ID | Category | Check | Risk if missing |
 |---|---|---|---|
-| A01 | 注入 | 参数化查询 | 数据泄露或丢失 |
-| A02 | 失效认证 | bcrypt + session rotation | 账号劫持 |
-| A03 | 敏感数据暴露 | 传输加密+存储加密+不泄漏 | 数据泄露 |
-| A04 | XXE | 禁用外部实体解析 | XML 注入 |
-| A05 | 失效访问控制 | 每次请求重新验权 | 越权操作 |
-| A06 | 安全配置错误 | 默认安全+无调试头 | 信息泄漏 |
-| A07 | XSS | 输入输出编码+CSP | 浏览器端代码执行 |
-| A08 | 不安全反序列化 | 不反序列化不可信数据 | RCE |
-| A09 | 已知漏洞组件 | 定期 audit | 已知 exploit |
-| A10 | 未授权重放 | nonce + rate limit | 自动化滥用 |
-
-## 参见
-
-- `spec/contracts/error-output-as-data.md` — 防御 prompt injection
-- `skills/test-first/SKILL.md` — 安全回归测试证据
+| A01 | Injection | Parameterized queries | Data disclosure or loss |
+| A02 | Broken authentication | bcrypt + session rotation | Account takeover |
+| A03 | Sensitive data exposure | Encryption in transit + at rest + no leakage | Data disclosure |
+| A04 | XXE | Disable external entity parsing | XML injection |
+| A05 | Broken access control | Re-authorize on every request | Unauthorized operations |
+| A06 | Security misconfiguration | Secure defaults + no debug headers | Information disclosure |
+| A07 | XSS | Input/output encoding + CSP | Browser-side code execution |
+| A08 | Insecure deserialization | Do not deserialize untrusted data | RCE |
+| A09 | Vulnerable components | Regular audits | Known exploits |
+| A10 | Unauthorized replay | Nonce + rate limit | Automated abuse |
+
+## See also
+
+- `spec/contracts/error-output-as-data.md` — defending against prompt injection
+- `skills/test-first/SKILL.md` — security regression test evidence

package/template/.cowork-flow/spec/references/testing-checklist.md

@@ -1,45 +1,45 @@
 # Testing Checklist
 
-> 测试质量快速参考。审查测试文件、编写测试用例或检查行为覆盖时按需加载。
+> Quick reference for test quality. Load on demand when reviewing test files, writing test cases, or checking behavior coverage.
 
-## 测试结构检查
+## Test structure
 
-- 测试名称描述行为：`test_<unit>_<scenario>_<expected>`
-- AAA 排列（Arrange/Act/Assert 之间有视觉分隔）
-- 一个测试只验证一个行为
-- 断言行为契约，而非实现细节
-- mock 验证行为调用，而非 call-count
+- Test names describe behavior: `test_<unit>_<scenario>_<expected>`
+- AAA layout (visual separation between Arrange/Act/Assert)
+- One test verifies one behavior
+- Assert behavior contracts, not implementation details
+- Mocks verify behavior calls, not call counts
 
-## 深度检查
+## Depth
 
-- 不是 `assert True` 通过的测试
-- 不是"只检查函数是否存在"的测试
-- 不是通过 `unittest.mock` 调用计数代替行为验证的测试
-- 不是快照空结构（`assert result == {}`）的测试
-- 不依赖测试执行顺序
-- 不跳过必要的 setup/teardown
-- 断言精确值或行为契约，而非 truthy/falsy
-- 异常路径有至少一个测试覆盖
+- Not a test that passes via `assert True`
+- Not a test that only checks whether a function exists
+- Not a test that uses `unittest.mock` call counts instead of behavior verification
+- Not a test that snapshots an empty structure (`assert result == {}`)
+- Does not depend on test execution order
+- Does not skip necessary setup/teardown
+- Asserts exact values or behavior contracts, not truthy/falsy
+- Has at least one test for the error path
 
-## 反模式
+## Anti-patterns
 
-| 反模式 | 问题 | 替代方案 |
+| Anti-pattern | Problem | Alternative |
 |---|---|---|
-| 设置期望值为硬编码输出 | 测试与实现耦合，实现变更但行为相同时测试失败 | 验证行为契约：输入 -> 期望输出关系 |
-| mock 受测函数的内部调用 | 测试实现而非行为 | mock 留在边界层（API/文件系统/时间） |
-| 多个不相关行为塞一个测试 | 失败时不知道是什么坏了 | 每行为独立测试，名称描述行为 |
-| 断言 `assert True` | 总是通过 | 改为语义断言（assertEqual/assertGreater 等） |
-| 只检查函数存在 `assert module.func` | 不验证行为 | 调用函数验输出或异常 |
-| 不清理全局状态 / fixtures | 测试间污染导致顺序依赖 | 使用 setup/teardown 清理 |
+| Hard-coding expected values to implementation output | Test couples to implementation; it fails when implementation changes but behavior does not | Verify the behavior contract: input -> expected output |
+| Mocking internal calls of the function under test | Tests implementation instead of behavior | Keep mocks at boundaries (API/filesystem/time) |
+| Packing unrelated behaviors into one test | A failure does not reveal what broke | One test per behavior, with a descriptive name |
+| Asserting `assert True` | Always passes | Use semantic assertions (assertEqual/assertGreater, etc.) |
+| Only checking that a function exists (`assert module.func`) | Does not verify behavior | Call the function and verify output or exceptions |
+| Not cleaning up global state / fixtures | Cross-test pollution causes order dependence | Clean up in setup/teardown |
 
-## 范围
+## Scope
 
-- 当前修改的函数/模块有直接单元测试
-- 下游依赖有集成测试
-- 边界条件（空值/极值/并发）有至少一个测试覆盖
+- The changed function/module has a direct unit test
+- Downstream dependencies have integration tests
+- Boundary conditions (empty/extreme values/concurrency) have at least one test
 
-## 参见
+## See also
 
-- `spec/backend/quality-guidelines.md` — 后端测试规范
-- `skills/test-first/SKILL.md` — 完整的 TDD 流程
-- `references/security-checklist.md` — 测试中的安全边界检查
+- `spec/backend/quality-guidelines.md` — backend testing rules
+- `skills/test-first/SKILL.md` — full TDD workflow
+- `references/security-checklist.md` — security boundary checks in tests

package/template/.cowork-flow/spec/runtime/host-assets.json

@@ -442,16 +442,36 @@
   ],
   "excludedPrefixes": [
     "skills/",
+    "locales/",
     ".zcode/",
     ".qoder/"
   ],
+  "localization": {
+    "default": "zh-CN",
+    "supported": [
+      "zh-CN",
+      "en-US"
+    ],
+    "assets": {
+      "en-US": [
+        ".cowork-flow/spec/contracts/decision-anchor.md",
+        ".cowork-flow/templates/plan-template.md"
+      ]
+    }
+  },
   "syncPolicy": {
     "protectedFiles": [
       ".cowork-flow/config.yaml",
       ".cowork-flow/.developer"
     ],
+    "forceProtectedFiles": [
+      ".cowork-flow/config.yaml"
+    ],
     "protectedPrefixes": [
-      ".cowork-flow/spec/",
+      ".cowork-flow/spec/backend/",
+      ".cowork-flow/spec/frontend/",
+      ".cowork-flow/spec/guides/",
+      ".cowork-flow/spec/references/",
       ".cowork-flow/tasks/",
       ".cowork-flow/plans/"
     ],
@@ -459,9 +479,7 @@
       ".cowork-flow/.gitignore",
       ".cowork-flow/.version",
       ".cowork-flow/run",
-      ".cowork-flow/run.cmd",
-      ".cowork-flow/spec/contracts/workflow-state-templates.md",
-      ".cowork-flow/spec/runtime/scope-rules.json"
+      ".cowork-flow/run.cmd"
     ],
     "safePrefixes": [
       ".cowork-flow/"
@@ -511,6 +529,7 @@
       ".claude/skills/update-spec",
       ".claude/skills/using-cowork-flow",
       ".claude/skills/writing-plans",
+      ".cowork-flow/plan-template.md",
       ".cowork-flow/project-context.md",
       ".cowork-flow/scripts/application",
       ".cowork-flow/scripts/services/batch_execution.py",

package/template/.cowork-flow/spec/runtime/index.md

@@ -1,19 +1,17 @@
 # Runtime Specs
 
-本目录存放 workflow runtime 直接读取的机器配置。
+This directory holds machine configuration read directly by the workflow runtime.
 
-- `contract-registry.json`: hook/plugin 注入 contract digest 的注册表。
-- `host-assets.json`: 宿主平台、资产归属、技能读取根与宿主发现通道（`skillReadRoot` / `skillDiscovery[]`，条目为仓库内路径加 `gates` / `evidence`；机器级载荷不作为技能来源参与）、机器级插件载荷声明（`payload`：载荷在包内的目录与插件清单名，无插件载荷的宿主为 `null`；机器安装位置不在此声明）、同步保护策略和旧资产迁移清单的单源；结构由 `../schemas/host-assets.schema.json` 约束。`capabilityMatrix.hosts` 必须覆盖 `platforms` 声明的每一个 id——必需宿主集合由平台数据派生，运行时不维护宿主名单常量。
+- `contract-registry.json`: registry of the contract digests injected by hooks/plugins.
+- `host-assets.json`: single source for host platforms, asset ownership, skill read roots and host discovery channels (`skillReadRoot` / `skillDiscovery[]`, entries are repo-relative paths plus `gates` / `evidence`; machine-level payloads do not participate as skill sources), machine-level plugin payload declarations (`payload`: the package-relative payload directory and plugin manifest name; `null` for hosts without a plugin payload; machine install locations are not declared here), the sync protection policy, and the obsolete-asset migration list. The structure is constrained by `../schemas/host-assets.schema.json`. `capabilityMatrix.hosts` must cover every id declared in `platforms` — the required host set is derived from platform data; the runtime does not maintain a host-name constant.
 
-缺少或损坏的 runtime 文件不能静默放行关键流程能力。宿主插件可使用最小 fallback
-避免崩溃，但必须在 digest 中暴露 warning；doctor/tests 负责发现缺失。
+Missing or corrupted runtime files must not silently grant critical workflow capabilities. Host plugins may use a minimal fallback to avoid crashing, but must expose a warning in the digest; doctor/tests are responsible for detecting absence.
 
-用户自然语言规范不在本目录注册；`task-review` skill 直接读取
-`spec/backend/`、`spec/frontend/`、`spec/guides/` 并输出审查结论。
+User natural-language specs are not registered in this directory; the `task-review` skill reads `spec/backend/`, `spec/frontend/`, and `spec/guides/` directly and reports review findings.
 
-运行时写入边界：
+Runtime write boundaries:
 
-- CLI/宿主/Git 适配层位于 `scripts/adapters/`，用例服务位于 `scripts/services/`。
-- JSON 状态通过 `scripts/infra/storage/` 读写并显式使用 UTF-8。
-- init/sync 通过 Asset Plan、staging、备份和 rollback 提交资产，版本文件最后更新。
-- 旧状态只允许在带迁移测试的读取边界兼容；新写入必须使用当前 schema 和权威路径。
+- CLI/host/Git adaptation lives in `scripts/adapters/`; use-case services live in `scripts/services/`.
+- JSON state is read and written through `scripts/infra/storage/` with explicit UTF-8.
+- init/sync commit assets through Asset Plan, staging, backup, and rollback; the version file is updated last.
+- Old state is only allowed at read boundaries with migration tests; new writes must use the current schema and authoritative paths.

package/template/.cowork-flow/spec/schemas/host-assets.schema.json

@@ -56,6 +56,9 @@
     "excludedPrefixes": {
       "$ref": "#/$defs/stringList"
     },
+    "localization": {
+      "$ref": "#/$defs/localization"
+    },
     "syncPolicy": {
       "$ref": "#/$defs/syncPolicy"
     }
@@ -68,6 +71,31 @@
         "minLength": 1
       }
     },
+    "localization": {
+      "description": "Per-language asset overrides. Paths are canonical project-relative paths; the base template is the default language.",
+      "type": "object",
+      "additionalProperties": false,
+      "required": [
+        "default",
+        "supported",
+        "assets"
+      ],
+      "properties": {
+        "default": {
+          "type": "string",
+          "minLength": 1
+        },
+        "supported": {
+          "$ref": "#/$defs/stringList"
+        },
+        "assets": {
+          "type": "object",
+          "additionalProperties": {
+            "$ref": "#/$defs/stringList"
+          }
+        }
+      }
+    },
     "capabilities": {
       "type": "object",
       "minProperties": 1,
@@ -342,6 +370,9 @@
         "protectedFiles": {
           "$ref": "#/$defs/stringList"
         },
+        "forceProtectedFiles": {
+          "$ref": "#/$defs/stringList"
+        },
         "protectedPrefixes": {
           "$ref": "#/$defs/stringList"
         },

package/template/.cowork-flow/spec/schemas/index.md

@@ -1,9 +1,9 @@
 # Schema Specs
 
-本目录存放 JSON Schema。
+This directory holds JSON Schemas.
 
-- `adapter.schema.json`: 宿主适配器声明 schema。
-- `host-assets.schema.json`: Host Asset Manifest schema（`runtime/host-assets.json` 的结构约束）。
-- `party-mode-v2-actions.schema.json`: Party Mode V2 action schema。
+- `adapter.schema.json`: host adapter declaration schema.
+- `host-assets.schema.json`: Host Asset Manifest schema (structural constraints for `runtime/host-assets.json`).
+- `party-mode-v2-actions.schema.json`: Party Mode V2 action schema.
 
-Schema 只描述数据合同，不承载流程说明；流程和行为边界写入 `contracts/`、Skill command manifests 或 `AGENTS.md`。
+Schemas describe data contracts only, not workflow guidance; workflow and behavior boundaries belong in `contracts/`, Skill command manifests, or `AGENTS.md`.

package/template/.cowork-flow/templates/plan-template.md

@@ -0,0 +1,91 @@
+# <功能> 实施计划
+
+> 执行规则：本计划用于防止实现漂移，实现者必须按任务顺序推进。
+> 如果计划与当前代码事实冲突、需要计划外文件，或测试因未计划的原因失败，停止并报告，不要扩大范围。
+>
+> 正式固定代理工作：使用 `.cowork-flow/run subagent init` 创建 runtime context，通过当前 Host Adapter 传入 `cowork_runtime_context_id: <runtime_context_id>`，再派发 `cowork-implement` 或 `cowork-check`；验证完成后关闭 runtime context。
+
+| 字段 | 内容 |
+|----|------|
+| **任务类型** | Tiny / Normal / High-risk：<理由> |
+| **执行策略** | Serial / Parallel：<理由> |
+| **成功标准** | AC-001 / AC-002 / AC-003 |
+| **最终验证** | `<command>` → exit 0；逐条核对适用的 `.cowork-flow/spec/<dev_type>/` 规范与项目门禁 → 每条 `pass` / `not_applicable` |
+| **主要风险** | Low / Medium / High：<一句话> |
+
+**目标:** <一句话描述可观察的行为变化>
+
+## 目标
+
+<1-3 句话描述预期可观察行为。>
+
+## 验收标准
+
+- AC-001: <可验证行为>
+- AC-002: <可验证行为>
+
+## 全局约束
+
+- <必须保留的架构边界、兼容要求、命令或输出字段>
+- <明确禁止的范围扩张>
+
+## 文件边界
+
+- `path/a.py`：<职责；只改这个函数或区域>
+- `tests/test_a.py`：<职责；只新增或修改这些测试>
+
+## 任务
+
+Tiny 任务可保留简化 brief，但仍须写明精确文件边界、完成条件和一条验证命令。
+
+### 任务 N：<小目标>
+
+**目的**
+- <本任务完成后的可观察行为>
+
+**代码事实来源**
+- Read: `path/a.py::function_name`
+- Read: `tests/test_a.py::test_existing_behavior`
+
+**文件边界**
+- Create / Modify / Test: `path/a.py`（只改 <函数/区域>）
+- Test: `tests/test_a.py`（只改 <测试/区域>）
+
+**关键符号**
+- Change: `function_name()`、`ClassName.method()`、`<config-key>`
+- Preserve: `existing_helper()`、输出字段 `<field>`
+
+**实现说明**
+```python
+# 核心控制流草图；按现有代码调整名称。
+# 不要把完整实现复制进计划。
+...
+```
+
+**测试证明**
+- `tests/test_a.py::test_x`：证明 <具体行为>
+- 核心断言：<一句话描述断言，不只有测试名>
+
+**验证命令**
+```bash
+<command>
+```
+
+当改动文件受 `.cowork-flow/spec/<dev_type>/` 规则覆盖时，在这里加入决定命令或规范门禁核对行（`pass` / `not_applicable` 并附理由）；这是强复核提醒，不是硬门禁。
+
+**完成条件**
+- <2-4 条可检查条件>
+
+**禁止漂移**
+- <2-5 条明确禁止的改动>
+
+**偏差条件**
+- 如果 <计划外文件/层> 变得必要，必须停止并报告，因为文件边界有误。
+
+## 最终验证
+
+```bash
+<command>
+```
+
+规范核对：逐条核对 `.cowork-flow/spec/<dev_type>/` 适用规范 → <pass / not_applicable + 理由>。