dominds 0.8.8 → 0.8.9

package/dist/docs/team-mgmt-toolset.md

@@ -1,6 +1,6 @@
-# Team Management Toolset (`team_mgmt`)
+# Team Management Toolset (`team-mgmt`)
 
-Chinese version: [中文版](./team_mgmt-toolset.zh.md)
+Chinese version: [中文版](./team-mgmt-toolset.zh.md)
 
 This document specifies a dedicated **team management toolset** whose only job is managing the
 rtws’s “mindset” configuration files under `.minds/` (team roster, LLM providers, and agent
@@ -29,7 +29,7 @@ equivalent of the `ws_mod` toolset + unrestricted `read_dirs`/`write_dirs`), bec
 
 ## Migration Plan (Replacing legacy builtin team-manager knowledge)
 
-This document is a **design spec** for the new `team_mgmt` toolset. It is not something we should
+This document is a **design spec** for the new `team-mgmt` toolset. It is not something we should
 ever tell an agent to “look up” at runtime.
 
 Instead, the runtime “single source of truth” for team management guidance should be
@@ -75,9 +75,9 @@ This makes sense for “normal” agents, but it blocks the team manager from do
 - Providing general-purpose file editing across the repo.
 - Making `.minds/` broadly writable by default team members.
 
-## Proposed `team_mgmt` Toolset
+## Proposed `team-mgmt` Toolset
 
-The `team_mgmt` toolset mirrors a minimal subset of `fs`/`txt`, but **hard-scopes** all operations to
+The `team-mgmt` toolset mirrors a minimal subset of `fs`/`txt`, but **hard-scopes** all operations to
 `.minds/` and rejects anything outside.
 
 ### Naming Conventions (Human / UI)
@@ -124,7 +124,7 @@ Notes:
   - Reject paths containing `..`.
   - Reject any path that resolves outside `.minds/` after normalization.
 - Prefer an explicit allowlist over “anything in the rtws”.
-  - For `team_mgmt`, that explicit allowlist is `.minds/**` (including `.minds/memory/**`) so the
+  - For `team-mgmt`, that explicit allowlist is `.minds/**` (including `.minds/memory/**`) so the
     team manager can repair accidental corruptions made by other tools (even though `.minds/memory/**`
     already has dedicated `memory` / `team_memory` tools for normal use).
 - Require explicit `.minds/...` paths and validate them; do not support “implicitly scoped” paths
@@ -133,7 +133,7 @@ Notes:
 ### Why a dedicated toolset (instead of only `read_dirs` / `write_dirs`)?
 
 `read_dirs` / `write_dirs` are still valuable, but they are configured in `.minds/team.yaml`, which
-may not exist during bootstrap. A dedicated `team_mgmt` toolset:
+may not exist during bootstrap. A dedicated `team-mgmt` toolset:
 
 - Lets the team manager create `.minds/team.yaml` safely from “zero state”.
 - Keeps the scope bounded even if the member’s directory allow/deny lists are empty.
@@ -387,7 +387,7 @@ member_defaults:
     - ws_read
     - memory
   # Default posture: deny `.minds/` edits for normal members.
-  # (Team management should be done via `team_mgmt` tools, not general file tools.)
+  # (Team management should be done via `team-mgmt` tools, not general file tools.)
   no_read_dirs:
     - .minds/team.yaml
     - .minds/llm.yaml
@@ -426,7 +426,7 @@ Best practices:
 - Keep `.minds/team.yaml` ownership tight; only the team manager should be able to edit it.
 - Avoid repeating built-in constraints in `team.yaml`:
   - `*.tsk/**` (encapsulated Taskdocs) are hard-denied for all general file tools.
-  - `.minds/**` is hard-denied for general file tools; only the dedicated `team_mgmt` toolset can access it.
+  - `.minds/**` is hard-denied for general file tools; only the dedicated `team-mgmt` toolset can access it.
     Put these in `no_*` only when you need extra explicitness; they are enforced regardless.
 
 ## Managing `.minds/team/<member>/*.md` (agent minds)
@@ -460,8 +460,8 @@ Suggested structure:
 
 Preferred behavior for initial bootstrap:
 
-- The shadow `fuxi` instance should get `team_mgmt` (and the manual tool), not broad `ws_mod`.
-- The shadow `pangu` instance should get broad rtws toolsets (e.g. `ws_read`, `ws_mod`, `os`), but not `team_mgmt`.
+- The shadow `fuxi` instance should get `team-mgmt` (and the manual tool), not broad `ws_mod`.
+- The shadow `pangu` instance should get broad rtws toolsets (e.g. `ws_read`, `ws_mod`, `os`), but not `team-mgmt`.
 - After `.minds/team.yaml` is created, the team definition becomes the source of truth.
 
 This avoids needing to grant full rtws access to configure the team.
@@ -472,7 +472,7 @@ This avoids needing to grant full rtws access to configure the team.
   `member_defaults.model`.
 - **Provider not found**: Ensure `.minds/team.yaml` `provider` keys exist in merged provider config
   (`dominds/main/llm/defaults.yaml` + `.minds/llm.yaml`).
-- **Access denied when editing `.minds/`**: Intended for general file tools; use `team_mgmt` tools.
+- **Access denied when editing `.minds/`**: Intended for general file tools; use `team-mgmt` tools.
 - **MCP tools not visible in Tools view**:
   - Confirm `.minds/mcp.yaml` exists and is valid.
   - Open **Problems** and look for MCP-related errors.

package/dist/docs/team-mgmt-toolset.zh.md

@@ -1,6 +1,6 @@
-# 团队管理工具集（`team_mgmt`）
+# 团队管理工具集（`team-mgmt`）
 
-英文版：[English](./team_mgmt-toolset.md)
+英文版：[English](./team-mgmt-toolset.md)
 
 本文档指定了一个专用的**团队管理工具集**，其唯一职责是管理 rtws（运行时工作区）`.minds/` 下的"心智"配置文件（团队名单、LLM 提供商和智能体心智文件），而不授予广泛的运行时工作区访问权限。
 
@@ -23,7 +23,7 @@
 
 ## 迁移计划（替换传统的内置团队管理者知识）
 
-本文档是新的 `team_mgmt` 工具集的**设计规范**。这不是我们应该在运行时让智能体"查阅"的内容。
+本文档是新的 `team-mgmt` 工具集的**设计规范**。这不是我们应该在运行时让智能体"查阅"的内容。
 
 相反，运行时团队管理的"单一事实来源"应该是函数工具 `team_mgmt_manual` 的输出。
 
@@ -61,9 +61,9 @@
 - 提供跨仓库的通用文件编辑
 - 默认让 `.minds/` 可广泛写入
 
-## 提议的 `team_mgmt` 工具集
+## 提议的 `team-mgmt` 工具集
 
-`team_mgmt` 工具集镜像 `fs`/`txt` 的最小子集，但**硬作用域**所有操作到 `.minds/` 并拒绝任何外部操作。
+`team-mgmt` 工具集镜像 `fs`/`txt` 的最小子集，但**硬作用域**所有操作到 `.minds/` 并拒绝任何外部操作。
 
 ### 命名约定（人类 / UI）
 
@@ -104,12 +104,12 @@
   - 拒绝包含 `..` 的路径
   - 拒绝规范化后解析到 `.minds/` 之外的任何路径
 - 优先使用显式白名单而非" rtws 中的任何内容"
-  - 对于 `team_mgmt`，该显式白名单是 `.minds/**`（包括 `.minds/memory/**`），以便团队管理者可以修复其他工具造成的意外损坏（即使 `.minds/memory/**` 已有专用的 `memory` / `team_memory` 工具供正常使用）
+  - 对于 `team-mgmt`，该显式白名单是 `.minds/**`（包括 `.minds/memory/**`），以便团队管理者可以修复其他工具造成的意外损坏（即使 `.minds/memory/**` 已有专用的 `memory` / `team_memory` 工具供正常使用）
 - 需要显式的 `.minds/...` 路径并验证它们；不支持像 `team.yaml` 这样的"隐式作用域"路径
 
 ### 为什么需要专用工具集（而不是仅 `read_dirs` / `write_dirs`）？
 
-`read_dirs` / `write_dirs` 仍然很有价值，但它们配置在 `.minds/team.yaml` 中，在引导期间可能不存在。专用的 `team_mgmt` 工具集：
+`read_dirs` / `write_dirs` 仍然很有价值，但它们配置在 `.minds/team.yaml` 中，在引导期间可能不存在。专用的 `team-mgmt` 工具集：
 
 - 让团队管理者能够从"零状态"安全地创建 `.minds/team.yaml`
 - 即使成员的目录允许/拒绝列表为空，也保持作用域边界
@@ -335,7 +335,7 @@ member_defaults:
     - ws_read
     - memory
   # 默认姿态：拒绝普通成员的 `.minds/` 编辑
-  #（团队管理应通过 `team_mgmt` 工具完成，而非通用文件工具）
+  #（团队管理应通过 `team-mgmt` 工具完成，而非通用文件工具）
   no_read_dirs:
     - .minds/team.yaml
     - .minds/llm.yaml
@@ -372,7 +372,7 @@ members:
 - 保持 `.minds/team.yaml` 的所有权严格；只有团队管理者应该能够编辑它
 - 避免在 `team.yaml` 中重复内置约束：
   - `*.tsk/**`（封装的 Taskdocs）对所有通用文件工具被硬性拒绝
-  - `.minds/**` 对通用文件工具被硬性拒绝；只有专用的 `team_mgmt` 工具集可以访问它
+  - `.minds/**` 对通用文件工具被硬性拒绝；只有专用的 `team-mgmt` 工具集可以访问它
     - 只有当你需要额外的显式性时才将这些放入 `no_*`；无论如何都会强制执行
 
 ## 管理 `.minds/team/<member>/*.md`（智能体心智）
@@ -406,8 +406,8 @@ members:
 
 初始引导的首选行为：
 
-- 影子成员 `fuxi` 实例应该获得 `team_mgmt`（和手册工具），而不是广泛的 `ws_mod`
-- 影子成员 `pangu` 实例应该获得广泛的 rtws 工具集（例如 `ws_read`、`ws_mod`、`os`），但不获得 `team_mgmt`
+- 影子成员 `fuxi` 实例应该获得 `team-mgmt`（和手册工具），而不是广泛的 `ws_mod`
+- 影子成员 `pangu` 实例应该获得广泛的 rtws 工具集（例如 `ws_read`、`ws_mod`、`os`），但不获得 `team-mgmt`
 - 在创建 `.minds/team.yaml` 后，团队定义成为事实来源
 
 这避免了需要授予完整的 rtws 访问权限来配置团队。
@@ -416,7 +416,7 @@ members:
 
 - **"缺少必需的 provider/model"**：确保 `.minds/team.yaml` 有 `member_defaults.provider` 和 `member_defaults.model`
 - **找不到提供商**：确保 `.minds/team.yaml` 的 `provider` 键存在于合并的提供商配置中（`dominds/main/llm/defaults.yaml` + `.minds/llm.yaml`）
-- **编辑 `.minds/` 时访问被拒绝**：通用文件工具的预期行为；使用 `team_mgmt` 工具
+- **编辑 `.minds/` 时访问被拒绝**：通用文件工具的预期行为；使用 `team-mgmt` 工具
 - **MCP 工具在工具视图中不可见**：
   - 确认 `.minds/mcp.yaml` 存在且有效
   - 打开**问题**并查找 MCP 相关错误

package/dist/minds/promptdocs.js

@@ -0,0 +1,263 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.buildNoToolsNotice = buildNoToolsNotice;
+exports.buildShellPolicyPrompt = buildShellPolicyPrompt;
+exports.buildMemorySystemPrompt = buildMemorySystemPrompt;
+exports.buildIntrinsicToolUsageText = buildIntrinsicToolUsageText;
+const minds_i18n_1 = require("./minds-i18n");
+function buildNoToolsNotice(language) {
+    return language === 'zh'
+        ? '无工具：不能调用任何工具，也不能访问 rtws / 文件 / 浏览器 / shell。'
+        : 'No tools: do not call any tools, and do not access the rtws, files, browser, or shell.';
+}
+function buildShellPolicyPrompt(ctx) {
+    const { language, agentIsShellSpecialist, agentHasShellTools, agentHasReadonlyShell, shellSpecialistMemberIds, } = ctx;
+    const title = language === 'zh' ? '### Shell 执行策略（重要）' : '### Shell Execution Policy (Important)';
+    const devopsScriptPolicy = language === 'zh'
+        ? 'DevOps 场景：忌讳临时脚本。不要用临时脚本完成工作；如需工具脚本，请与队友和人类讨论并在 rtws 中正式设计/命名/规范化后使用。'
+        : 'DevOps context: ad-hoc temp scripts are a taboo. Do not rely on temp scripts; if a tool script is needed, align with teammates and the human and formalize it in the rtws before use.';
+    const shellSpecialists = shellSpecialistMemberIds.length > 0
+        ? shellSpecialistMemberIds.map((id) => `@${id}`).join(', ')
+        : language === 'zh'
+            ? '（未配置 shell 专员）'
+            : '(no shell specialists configured)';
+    if (agentIsShellSpecialist && agentHasShellTools) {
+        const body = language === 'zh'
+            ? [
+                '你是本队的 shell 专员，具备 shell 执行能力（高风险）。使用前必须先做风险识别与最小化：',
+                '- 说明目的、预期输出/验证方式、预期工作目录与影响面。',
+                '- 优先选择只读命令；写入/删除/网络/权限相关操作必须解释理由与安全边界。',
+                '- 如果不确定命令后果，先提出更安全的替代方案或向人类/队友确认。',
+                '',
+                '当队友请求你执行命令时：请先复述你将执行的命令与风险边界，再执行，并按结构化格式回传 command/exit_code/stdout/stderr。',
+                '',
+                `本队 shell 专员列表：${shellSpecialists}`,
+            ].join('\n')
+            : [
+                'You are a designated shell specialist and have shell execution capability (high-risk). Before using it, do risk identification and minimize blast radius:',
+                '- State intent, expected output/verification, expected working directory, and scope of impact.',
+                '- Prefer read-only commands; for writes/deletes/network/privilege-related actions, justify and state safety boundaries.',
+                '- If unsure about consequences, propose a safer alternative or confirm with a human/teammate first.',
+                '',
+                'When teammates ask you to run commands: restate the exact command and guardrails, then execute, and report back with a structured receipt: command/exit_code/stdout/stderr.',
+                '',
+                `Shell specialists in this team: ${shellSpecialists}`,
+                '',
+                devopsScriptPolicy,
+            ].join('\n');
+        return `${title}\n\n${body}`.trim();
+    }
+    if (agentIsShellSpecialist && !agentHasShellTools) {
+        const body = language === 'zh'
+            ? [
+                '你被配置为 shell 专员，但当前没有可用的 shell 工具（团队配置错误）。',
+                '在未修复配置前：不要声称自己执行过命令；如需验证，请让人类修复 team.yaml 或把 shell 能力授予正确的成员。',
+                '',
+                `本队 shell 专员列表：${shellSpecialists}`,
+            ].join('\n')
+            : [
+                'You are configured as a shell specialist, but you do not currently have shell tools available (team configuration error).',
+                'Until it is fixed: do not claim you ran any command; ask a human to fix team.yaml or grant shell tools to the correct specialist member.',
+                '',
+                `Shell specialists in this team: ${shellSpecialists}`,
+                '',
+                devopsScriptPolicy,
+            ].join('\n');
+        return `${title}\n\n${body}`.trim();
+    }
+    const body = language === 'zh'
+        ? agentHasReadonlyShell
+            ? [
+                '你不具备高权限 shell 工具（shell_cmd/stop_daemon/get_daemon_output）。',
+                '你已被明确授权使用 `readonly_shell` 自行执行只读命令（仅允许白名单命令前缀）：请直接调用，不要去寻找 shell 专员代跑。',
+                '',
+                '当你需要的命令不在白名单内，或需要写入/删除/网络/长时间运行/进程管理等高风险能力时：再转交给具备 shell 工具的专员队友，并提供充分理由与可审查的命令提案：',
+                '- 你要达成的目标（why）',
+                '- 建议命令（what）+ 预期工作目录（cwd）+ 预期输出/验证方式（how to verify）',
+                '- 风险评估与安全边界（risk & guardrails）',
+                `可转交的 shell 专员队友：${shellSpecialists}`,
+                '',
+                devopsScriptPolicy,
+                '',
+                '重要：如果你打算让队友执行命令，请在同一条消息里给出完整的 tellask 诉请块（以第 0 列开头的 `!?@<shell-specialist>` 行），不要只说“我会请某人运行”。',
+                '重要：在你看到 shell 专员的回执（command/exit_code/stdout/stderr）之前，不要声称“已运行/已通过/无错”。',
+            ].join('\n')
+            : [
+                '你不具备 shell 工具（本环境仅 shell 专员可执行 shell）：不要尝试“编造/假设”命令输出，也不要要求系统直接执行。',
+                '当你确实需要 shell 执行时：请转交给具备 shell 能力的专员队友，并提供充分理由与可审查的命令提案：',
+                '- 你要达成的目标（why）',
+                '- 建议命令（what）+ 预期工作目录（cwd）+ 预期输出/验证方式（how to verify）',
+                '- 风险评估与安全边界（risk & guardrails）',
+                `可转交的 shell 专员队友：${shellSpecialists}`,
+                '',
+                devopsScriptPolicy,
+                '',
+                '重要：如果你打算让队友执行命令，请在同一条消息里给出完整的 tellask 诉请块（以第 0 列开头的 `!?@<shell-specialist>` 行），不要只说“我会请某人运行”。',
+                '重要：在你看到 shell 专员的回执（command/exit_code/stdout/stderr）之前，不要声称“已运行/已通过/无错”。',
+            ].join('\n')
+        : agentHasReadonlyShell
+            ? [
+                'You do not have high-risk shell tools (shell_cmd/stop_daemon/get_daemon_output).',
+                'You are explicitly authorized to use `readonly_shell` yourself for read-only inspection via its small allowlist. Call it directly; do not go looking for a shell specialist to run it for you.',
+                '',
+                'When the command you need is not in the allowlist, or you need high-risk capabilities like writes/deletes/network/long-running jobs/process management: delegate to a shell specialist teammate with a justified, reviewable proposal:',
+                '- Goal (why)',
+                '- Proposed command (what) + expected working directory (cwd) + expected output/verification (how to verify)',
+                '- Risk assessment and guardrails (risk & guardrails)',
+                '',
+                `Shell specialist teammates: ${shellSpecialists}`,
+                '',
+                devopsScriptPolicy,
+                '',
+                'Important: if you intend to delegate, include the full tellask block (a column-0 `!?@<shell-specialist>` line) in the same message; do not just say “I will ask someone to run it”.',
+                'Important: do not claim “ran/passed/no errors” until you see the shell specialist’s receipt (command/exit_code/stdout/stderr).',
+            ].join('\n')
+            : [
+                'You do not have shell tools configured (shell execution is restricted to designated specialists): do not fabricate/assume command output, and do not ask the system to execute commands directly.',
+                'When you truly need shell execution, delegate to a shell specialist teammate with a justified, reviewable proposal:',
+                '- Goal (why)',
+                '- Proposed command (what) + expected working directory (cwd) + expected output/verification (how to verify)',
+                '- Risk assessment and guardrails (risk & guardrails)',
+                '',
+                `Shell specialist teammates: ${shellSpecialists}`,
+                '',
+                devopsScriptPolicy,
+                '',
+                'Important: if you intend to delegate, include the full tellask block (a column-0 `!?@<shell-specialist>` line) in the same message; do not just say “I will ask someone to run it”.',
+                'Important: do not claim “ran/passed/no errors” until you see the shell specialist’s receipt (command/exit_code/stdout/stderr).',
+            ].join('\n');
+    return `${title}\n\n${body}`.trim();
+}
+function buildMemorySystemPrompt(ctx) {
+    const { language } = ctx;
+    const title = language === 'zh' ? '### 记忆系统（重要）' : '### Memory System (Important)';
+    if (language === 'zh') {
+        const teamMemoryLine = '- 团队记忆：稳定的团队约定/工程规约（跨任务共享）。';
+        const personalMemoryLine = '- 个人记忆：稳定的个人习惯/偏好与职责域知识；可维护你职责范围内的“rtws 索引”（关键文档/代码的准确路径 + 必要要点），以减少重复读文件；不要记录具体任务状态。';
+        const body = [
+            '你的聊天记录与工具输出是临时信息：会快速累积、很快过时，并增加你的认知负担。在同一轮对话中，除了 `clear_mind` 以外你无法真正丢弃这些历史。',
+            '`clear_mind` 会开启新一程对话（保留差遣牒、提醒项与记忆层），从而卸掉这部分认知负载并继续推进。因此你必须先把关键信息提炼到高价值载体：',
+            '- 差遣牒（Taskdoc，`*.tsk/`）：全队共享的任务契约（goals/constraints/progress）；保持足够短，每轮都应可通读。',
+            `- 更新差遣牒的任意分段时：每次调用会替换该分段全文；你必须先对照“上下文中注入的当前内容”做合并/压缩；禁止覆盖/抹掉他人条目；自己负责维护的条目必须标注责任人（例如 \`- [owner:@${ctx.agentId}] ...\` 或用 \`### @${ctx.agentId}\` 分块）。`,
+            '- 其中 `progress` 是全队共享公告牌：用于“阶段性进度快照”（关键决策/当前状态/下一步），不是流水账。',
+            '- 重要：差遣牒内容会被系统以内联形式注入到上下文中（本轮生成视角下即为最新）。需要回顾时请直接基于上下文里的差遣牒内容回顾与决策，不要试图用通用文件工具读取 `*.tsk/` 下的文件（会被拒绝）。',
+            '- 约定：`constraints` 尽量只写任务特有的硬要求。系统提示/工具文档里已明确且由系统强制执行的通用规则（例如 `*.tsk/` 封装禁止通用文件工具）无需重复写入 `constraints.md`。',
+            '- 提醒项（工作集）：当前对话的高频工作记录/关键细节（偏私有，不作为全队公告）；保持少量（常见 1–3 条），优先 `update_reminder` 压缩/合并，不再需要就 `delete_reminder`。',
+            teamMemoryLine,
+            personalMemoryLine,
+            '',
+            ...(ctx.isSubdialog
+                ? [
+                    `你当前处于支线对话：此处不允许 \`change_mind\`。当你判断需要更新差遣牒（尤其是 progress 公告牌）时，请在合适时机直接诉请差遣牒维护人 \`@${ctx.taskdocMaintainerId}\` 执行更新，并给出你已合并好的“新全文/替换稿”（用于替换对应章节全文）。不要声称已更新，除非看到回执。`,
+                ]
+                : [
+                    '你当前处于主线对话：你负责综合维护全队共享差遣牒（尤其是 progress 公告牌）。当队友/支线对话提出更新建议时，及时合并、压缩并保持清晰。',
+                ]),
+            ...(ctx.agentHasTeamMemoryTools
+                ? [
+                    '提示：你具备团队记忆工具（`add_team_memory` / `replace_team_memory` / `drop_team_memory` / `clear_team_memory`），可在必要时维护团队记忆（谨慎、少量、只写稳定约定）。',
+                ]
+                : []),
+            ...(ctx.agentHasPersonalMemoryTools
+                ? [
+                    '提示：你具备个人记忆工具（`add_memory` / `replace_memory` / `drop_memory` / `clear_memory`）。目标：维护你职责域的“rtws 索引”（关键文档/代码的准确路径 + 最小必要要点，如入口文件/关键符号/约定），让你在职责范围内尽量做到“0 次 ripgrep 就能开始干活”；一旦你修改了相关文件或发现记忆有过期/冲突，必须立刻用 `replace_memory` 把对应条目更新为最新事实。',
+                ]
+                : []),
+            ...(ctx.isSubdialog
+                ? [
+                    `工作流：先做事 → 再提炼（\`update_reminder\`；必要时整理差遣牒更新提案并诉请 \`@${ctx.taskdocMaintainerId}\` 合并写入）→ 然后 \`clear_mind\` 清空噪音。`,
+                ]
+                : [
+                    '工作流：先做事 → 再提炼（`update_reminder` + `change_mind(progress)`）→ 然后 `clear_mind` 清空噪音。',
+                ]),
+            '当 context health 变黄/红：立刻停止继续大实现/大阅读；先提炼，再 clear。',
+            '不要把长日志/大段 tool output 直接塞进差遣牒；差遣牒只写结论+下一步；细节只保留必要摘录放提醒项。',
+        ].join('\n');
+        return `${title}\n\n${body}`.trim();
+    }
+    const teamMemoryLine = '- Team memory: stable shared conventions (cross-task).';
+    const personalMemoryLine = '- Personal memory: stable personal habits/preferences and responsibility-scope knowledge. Maintain a compact responsibility-area rtws index (exact key doc/code paths + minimal key facts) to reduce repeat file reads; do not store per-task state.';
+    const body = [
+        'Dialog history and tool outputs are temporary: they accumulate quickly, become stale, and increase cognitive load. Within a course, you cannot truly drop that history except via `clear_mind`.',
+        '`clear_mind` starts a new course while preserving Taskdoc, reminders, and memory layers. Therefore, before clearing, distill key information into durable layers:',
+        '- Taskdoc (`*.tsk/`): team-shared task contract (goals/constraints/progress). Keep it small enough to read every course.',
+        `- When updating any Taskdoc section: each call replaces the entire section; always start from the current injected content and merge/compress; do not overwrite other contributors; add an explicit owner tag for entries you maintain (e.g., \`- [owner:@${ctx.agentId}] ...\` or a \`### @${ctx.agentId}\` block).`,
+        '- Taskdoc `progress` is the team’s shared bulletin board: distilled milestone snapshots (key decisions/current status/next steps), not raw logs.',
+        '- Important: the Taskdoc content is injected inline into the context (the latest as of this generation). Review the injected Taskdoc instead of trying to read files under `*.tsk/` via general file tools (they will be rejected).',
+        '- Convention: keep Taskdoc `constraints` focused on task-specific requirements. Do not duplicate global, system-enforced rules already stated in system prompt/tool docs (e.g. `.tsk/` encapsulation bans general file tools).',
+        '- Reminders (working set): your high-frequency per-dialog worklog + critical details (not a team bulletin board); keep it small (often 1–3 items), prefer `update_reminder` to compress/merge; delete when obsolete.',
+        teamMemoryLine,
+        personalMemoryLine,
+        '',
+        ...(ctx.isSubdialog
+            ? [
+                `You are currently in a subdialog: \`change_mind\` is not allowed here. When Taskdoc should be updated (especially the shared progress bulletin board), tellask the Taskdoc maintainer \`@${ctx.taskdocMaintainerId}\` with a fully merged replacement draft (full-section replacement). Do not claim it is updated until you see a receipt.`,
+            ]
+            : [
+                'You are currently in the main dialog: you are responsible for keeping the team-shared Taskdoc coherent and up to date (especially the progress bulletin board). Merge proposals from teammates/subdialogs promptly and keep it concise.',
+            ]),
+        ...(ctx.agentHasTeamMemoryTools
+            ? [
+                'Hint: you have team-memory tools (`add_team_memory` / `replace_team_memory` / `drop_team_memory` / `clear_team_memory`) and may maintain team memory when it is truly stable and worth sharing.',
+            ]
+            : []),
+        ...(ctx.agentHasPersonalMemoryTools
+            ? [
+                'Hint: you have personal-memory tools (`add_memory` / `replace_memory` / `drop_memory` / `clear_memory`). Goal: maintain a compact responsibility-area rtws index (exact key doc/code paths + minimal key facts) so you can start work with 0 ripgrep within your scope. If you changed those files or detect staleness/conflicts, immediately `replace_memory` to keep it accurate.',
+            ]
+            : []),
+        ...(ctx.isSubdialog
+            ? [
+                `Workflow: do work → distill (\`update_reminder\`; when Taskdoc needs updates, draft a merged replacement and ask \`@${ctx.taskdocMaintainerId}\`) → then \`clear_mind\` to drop noise.`,
+            ]
+            : [
+                'Workflow: do work → distill (`update_reminder` + `change_mind(progress)`) → then `clear_mind` to drop noise.',
+            ]),
+        'When context health turns yellow/red: treat it as a hard stop; distill first, then clear.',
+        'Do not paste long logs/tool outputs into Taskdoc; Taskdoc should record decisions + next steps; keep only essential excerpts in reminders.',
+    ].join('\n');
+    return `${title}\n\n${body}`.trim();
+}
+function buildFuncToolUsageText(language, funcTools) {
+    if (funcTools.length === 0) {
+        return language === 'zh' ? '没有可用的函数工具。' : 'No function tools available.';
+    }
+    return funcTools
+        .map((tool) => {
+        const schema = tool.parameters;
+        const isRecord = (value) => typeof value === 'object' && value !== null && !Array.isArray(value);
+        const requiredValue = schema['required'];
+        const required = Array.isArray(requiredValue) && requiredValue.every((v) => typeof v === 'string')
+            ? requiredValue
+            : [];
+        const req = required.length > 0 ? required.join(', ') : (0, minds_i18n_1.noneRequiredFieldsText)(language);
+        const propsValue = schema['properties'];
+        const propsRecord = isRecord(propsValue) ? propsValue : {};
+        const props = Object.entries(propsRecord)
+            .map(([k, v]) => {
+            const desc = isRecord(v) && 'description' in v && typeof v.description === 'string'
+                ? v.description
+                : 'parameter';
+            return `- ${k}: ${desc}`;
+        })
+            .join('\n');
+        const labels = (0, minds_i18n_1.funcToolUsageLabels)(language);
+        let toolDesc = tool.descriptionI18n
+            ? (tool.descriptionI18n[language] ?? tool.descriptionI18n.en ?? tool.description)
+            : (tool.description ?? '');
+        if (tool.name === 'readonly_shell') {
+            toolDesc =
+                language === 'zh'
+                    ? '只读白名单 shell。node/python 仅允许版本探针，脚本执行一律拒绝。'
+                    : 'Read-only allowlist shell. Node/python version probes only; scripts are rejected.';
+        }
+        return `#### ${labels.toolLabel}: ${tool.name}\n\n${toolDesc}\n\n- ${labels.invocationLabel}: ${labels.invocationBody}\n- ${labels.requiredLabel}: ${req}\n- ${labels.parametersLabel}:\n${props}`.trim();
+    })
+        .join('\n\n');
+}
+function buildIntrinsicToolUsageText(language, funcTools) {
+    const usage = buildFuncToolUsageText(language, funcTools).trim();
+    if (usage === '')
+        return language === 'zh' ? '（无）' : '(none)';
+    return usage;
+}

package/dist/server/api-routes.js

@@ -45,9 +45,9 @@ exports.handleApiRoute = handleApiRoute;
 const promises_1 = __importDefault(require("fs/promises"));
 const path = __importStar(require("path"));
 const agent_priming_1 = require("../agent-priming");
-const dialog_run_state_1 = require("../dialog-run-state");
 const dialog_1 = require("../dialog");
 const dialog_global_registry_1 = require("../dialog-global-registry");
+const dialog_run_state_1 = require("../dialog-run-state");
 const log_1 = require("../log");
 const persistence_1 = require("../persistence");
 const diligence_1 = require("../shared/diligence");

package/dist/showing-by-doing.js

@@ -36,8 +36,11 @@ Object.defineProperty(exports, "__esModule", { value: true });
 exports.getShowingByDoingCacheStatus = getShowingByDoingCacheStatus;
 exports.scheduleShowingByDoingForNewDialog = scheduleShowingByDoingForNewDialog;
 const dialog_run_state_1 = require("./dialog-run-state");
+const client_1 = require("./llm/client");
 const driver_1 = require("./llm/driver");
+const registry_1 = require("./llm/gen/registry");
 const log_1 = require("./log");
+const load_1 = require("./minds/load");
 const runtime_language_1 = require("./shared/runtime-language");
 const id_1 = require("./shared/utils/id");
 const inter_dialog_format_1 = require("./shared/utils/inter-dialog-format");
@@ -555,11 +558,29 @@ async function generatePrimingNoteViaMainlineAgent(options) {
     const { dlg, snapshotCmd, snapshotText, fbrResponses, shellPolicy, fbrEffort } = options;
     const language = (0, runtime_language_1.getWorkLanguage)();
     try {
-        // Use the normal dialog driver (streaming/non-streaming selection, retries, etc.) with a
-        // toolless `@self` subdialog, so this behavior matches the runtime driver with no local drift.
-        //
-        // Tool-less is important here: the Agent Priming synthesis must be a pure text transform of
-        // snapshot + drafts, and must not run tools (including any mind-edit tools).
+        const minds = await (0, load_1.loadAgentMinds)(dlg.agentId, dlg);
+        const team = minds.team;
+        const agent = minds.agent;
+        const systemPrompt = minds.systemPrompt;
+        const provider = agent.provider ?? team.memberDefaults.provider;
+        const model = agent.model ?? team.memberDefaults.model;
+        if (!provider || !model) {
+            throw new Error(`Missing provider/model for agent '${dlg.agentId}'`);
+        }
+        const llmCfg = await client_1.LlmConfig.load();
+        const providerCfg = llmCfg.getProvider(provider);
+        if (!providerCfg) {
+            throw new Error(`Provider configuration not found: '${provider}'`);
+        }
+        const llmGen = (0, registry_1.getLlmGenerator)(providerCfg.apiType);
+        if (!llmGen) {
+            throw new Error(`LLM generator not found for apiType='${providerCfg.apiType}'`);
+        }
+        // Force non-streaming for this synthetic summary generation.
+        const agentForGen = Object.create(agent);
+        agentForGen.streaming = false;
+        // Intentionally pass zero tools: we want a pure text synthesis and must avoid accidental
+        // `clear_mind` / `change_mind` calls during dialog creation.
         const prompt = formatPrimingSynthesisPrompt({
             language,
             snapshotCmd,
@@ -568,35 +589,17 @@ async function generatePrimingNoteViaMainlineAgent(options) {
             fbrDrafts: fbrResponses,
             shellPolicy,
         });
-        const tellaskHead = '@self (agent priming synthesis)';
-        const tellaskBody = prompt;
-        const sub = await dlg.createSubDialog(dlg.agentId, tellaskHead, tellaskBody, {
-            originMemberId: dlg.agentId,
-            callerDialogId: dlg.id.selfId,
-            callId: (0, id_1.generateShortId)(),
-            collectiveTargets: [dlg.agentId],
-        });
-        const initPrompt = (0, inter_dialog_format_1.formatAssignmentFromSupdialog)({
-            fromAgentId: dlg.agentId,
-            toAgentId: sub.agentId,
-            tellaskHead,
-            tellaskBody,
-            language,
-            collectiveTargets: [dlg.agentId],
-        });
-        await (0, driver_1.driveDialogStream)(sub, { content: initPrompt, msgId: (0, id_1.generateShortId)(), grammar: 'markdown' }, true);
-        const saying = extractLastAssistantSaying(sub.msgs).trim();
-        if (saying) {
+        const ctxMsgs = [{ type: 'environment_msg', role: 'user', content: prompt }];
+        const result = await llmGen.genMoreMessages(providerCfg, agentForGen, systemPrompt, [], ctxMsgs, dlg.activeGenSeq, undefined);
+        const saying = result.messages
+            .filter((m) => m.type === 'saying_msg')
+            .map((m) => m.content)
+            .join('\n')
+            .trim();
+        if (saying)
             return saying;
-        }
     }
     catch (err) {
-        const message = err instanceof Error ? err.message : String(err);
-        // Do not silently fall back for known severe configuration errors (e.g. Codex requires
-        // streaming=true).
-        if (message.includes('apiType=codex requires streaming=true')) {
-            throw err;
-        }
         log_1.log.warn('Failed to generate Agent Priming note via mainline agent (best-effort)', err, {
             dialogId: dlg.id.valueOf(),
         });

package/dist/snippets/README.en.md

@@ -0,0 +1,3 @@
+# Prompt Templates
+
+This directory stores bilingual system prompt templates (Snippets).

package/dist/static/assets/{_baseUniq-D14YtKr6.js → _baseUniq-rwtkYcxR.js}

@@ -1,4 +1,4 @@
-import { aV as isObjectLike, br as baseGetTag, aE as isArray, aT as Symbol$1, bs as arrayLikeKeys, bt as baseKeys, aD as isArrayLike, bu as memoize, bv as isArguments, bw as MapCache, bm as eq, bx as Uint8Array, aW as getTag, a$ as isBuffer, b2 as Stack, by as isTypedArray, aZ as isObject, bz as isLength, bp as isIndex, aF as identity, bn as baseFor, bA as Set } from "./index-CUGcPcXL.js";
+import { aV as isObjectLike, br as baseGetTag, aE as isArray, aT as Symbol$1, bs as arrayLikeKeys, bt as baseKeys, aD as isArrayLike, bu as memoize, bv as isArguments, bw as MapCache, bm as eq, bx as Uint8Array, aW as getTag, a$ as isBuffer, b2 as Stack, by as isTypedArray, aZ as isObject, bz as isLength, bp as isIndex, aF as identity, bn as baseFor, bA as Set } from "./index-B2yxi4rE.js";
 var symbolTag$1 = "[object Symbol]";
 function isSymbol(value) {
   return typeof value == "symbol" || isObjectLike(value) && baseGetTag(value) == symbolTag$1;
@@ -658,4 +658,4 @@ export {
   hasIn as x,
   toString as y
 };
-//# sourceMappingURL=_baseUniq-D14YtKr6.js.map
+//# sourceMappingURL=_baseUniq-rwtkYcxR.js.map