feat: 创建与更新技能支持独立长期指令 (#211)

创建与更新技能增加可选 instructions 字符串：普通 Markdown 全文、省略保留、字符串整体替换、空字符串清空。内容归属明确为人格、原则、长期指令与技能方法；既有 AgentFS 直接编辑遵循实际权限与审批。create-agent 2.8.0 / update-agent 3.4.0 的最低客户端为 10.0.179，双语正文与 sidecar 同步。 Add optional plain-Markdown instructions to Agent creation and update guidance, including preserve/replace/clear semantics and clear content ownership. Existing direct AgentFS editing retains its permission and approval rules. Both localized bodies and catalog sidecars require DesireCore 10.0.179. 验证 / Validation：目录与 i18n 校验 0 errors；严格 sidecar 全覆盖检查 0 errors（既有证据警告保留）；双语由人工同步并按真实 source body 计算 freshness hash。依据用户明确授权，本次以全树凭据形状扫描、Git/PR 元数据扫描与客户中立性人工审阅代替缺失的私有敏感词表检查；凭据形状扫描 740 文件无命中，不声称执行私有词表匹配。
2026-10-21 06:24:08 +08:00 · 2026-10-07 21:03:40 -07:00
parent 8f6aa67c96
commit 13c72eba29
6 changed files with 54 additions and 32 deletions
--- a/skills/create-agent/SKILL.md
+++ b/skills/create-agent/SKILL.md
@@ -2,8 +2,8 @@
 license: "MIT"
 name: create-agent
 description: >-
-  通过多轮对话收集需求，调用 ManageAgent 内置工具创建新的 AgentFS v2 智能体，支持自定义 persona 和 principles。Use when 用户要求创建新智能体、培养某领域助手、或快速基于模板生成可治理 Agent。可选头像生成需配置图像服务，适用供应商条款/额度及独立费用。
-version: 2.7.0
+  通过多轮对话收集需求，调用 ManageAgent 内置工具创建新的 AgentFS v2 智能体，支持自定义 persona、principles 和可选 instructions。Use when 用户要求创建新智能体、培养某领域助手、或快速基于模板生成可治理 Agent。可选头像生成需配置图像服务，适用供应商条款/额度及独立费用。
+version: 2.8.0
 compatibility: "DesireCore ManageAgent; optional avatar generation requires a configured image provider under its own terms and quota, with separate charges; unavailable generation retains the existing avatar."
 type: meta
 risk_level: low
@@ -15,7 +15,7 @@ tags:
  - meta
 metadata:
  author: desirecore
-  updated_at: '2026-10-03'
+  updated_at: '2026-10-08'
  i18n:
    default_locale: en-US
    source_locale: zh-CN
@@ -26,17 +26,17 @@ metadata:
      name: 创建智能体
      short_desc: 通过自然语言对话收集需求，一键创建专业化数字智能体
      description: >-
-        通过多轮对话收集需求，调用 ManageAgent 内置工具创建新的 AgentFS v2 智能体，支持自定义 persona 和 principles。Use when 用户要求创建新智能体、培养某领域助手、或快速基于模板生成可治理 Agent。可选头像生成需配置图像服务，适用供应商条款/额度及独立费用。
+        通过多轮对话收集需求，调用 ManageAgent 内置工具创建新的 AgentFS v2 智能体，支持自定义 persona、principles 和可选 instructions。Use when 用户要求创建新智能体、培养某领域助手、或快速基于模板生成可治理 Agent。可选头像生成需配置图像服务，适用供应商条款/额度及独立费用。
      body: ./SKILL.zh-CN.md
-      source_hash: sha256:7ffb7a362c26a91e
+      source_hash: sha256:3c0613f2d9e9479a
      translated_by: human
    en-US:
      name: Create Agent
      short_desc: Collect requirements through natural-language conversation and create a specialized digital Agent in one step
      description: >-
-        Collect requirements through multi-turn conversation and call the ManageAgent builtin tool to create a new AgentFS v2 Agent, with customizable persona and principles. Use when the user asks to create a new Agent, raise a domain assistant, or quickly produce a governable Agent from a template. Optional avatar generation requires a configured image service under provider terms/quota and separate charges.
+        Collect requirements through multi-turn conversation and call the ManageAgent builtin tool to create a new AgentFS v2 Agent, with customizable persona, principles and optional standing instructions. Use when the user asks to create a new Agent, raise a domain assistant, or quickly produce a governable Agent from a template. Optional avatar generation requires a configured image service under provider terms/quota and separate charges.
      body: ./SKILL.md
-      source_hash: sha256:7ffb7a362c26a91e
+      source_hash: sha256:3c0613f2d9e9479a
      translated_by: human
 market:
  icon: >-
@@ -58,7 +58,7 @@ market:
    verified: true
  compatible_agents: []
  channel: latest
-  required_client_version: 10.0.108
+  required_client_version: 10.0.179
 ---

 # create-agent skill
@@ -69,7 +69,7 @@ Collect requirements through natural-language conversation and call the ManageAg

 ## L1: Overview

-Meta-skill: gather requirements over multi-turn conversation → generate persona/principles → land via `ManageAgent(action='create')`. Use it to raise a domain specialist (legal advisor, financial analyst), deploy a customized business Agent quickly, or produce a prototype from a template. The created repo conforms to the AgentFS v2 spec and is version-managed by git (governable, traceable). Its value is what the tool can't give: domain-tailored persona/principles generation + a pre-create preview confirmation.
+Meta-skill: gather requirements over multi-turn conversation → generate persona/principles and standing instructions → land via `ManageAgent(action='create')`. Use it to raise a domain specialist (legal advisor, financial analyst), deploy a customized business Agent quickly, or produce a prototype from a template. The created repo conforms to the AgentFS v2 spec and is version-managed by git (governable, traceable). Its value is what the tool can't give: domain-tailored persona/principles generation + a pre-create preview confirmation.

 ## L2: Detailed Spec

@@ -87,9 +87,10 @@ Trigger (any): user explicitly says "create an Agent / make me an assistant"; de

 ### Stage 3: Content Generation

-Organize persona and principles as structured fields (do not emit raw markdown; present field-by-field). Leave uncollected fields empty for the system to fill with defaults:
+Organize persona, principles and optional standing instructions as structured fields (do not emit raw markdown; present field-by-field). Leave uncollected fields empty for the system to fill with defaults:

 - **persona**: L0 one-sentence core identity; L1 `role` / `personality` (array of trait tags) / `communication_style`; L2 specialty, values, decision preferences (free-form).
+- **instructions**: optional plain Markdown string for stable responsibilities, default work strategies and delivery requirements; the full text is loaded without L0/L1/L2 parsing or truncation. Detailed procedures and reference material belong to skills. If only the name is known, omit this file.
 - **principles**: L0 one-sentence top principle; L1 `must_do` / `must_not` (safety red lines) / `priority`; L2 governance principles, escalation rules (free-form).

 **Domain matching reference** (recommend personality and red lines by domain):
@@ -103,7 +104,7 @@ Organize persona and principles as structured fields (do not emit raw markdown;

 ### Stage 4: User Confirmation

-Present the preview in natural language / tables (name, description, persona, principles; **no raw markdown source**), e.g.:
+Present the preview in natural language / tables (name, description, persona, principles, and optional standing instructions; **no raw markdown source**), e.g.:

 > About to create "Legal Advisor Assistant" — focused on contract review and legal risk assessment.
 > **Persona**: digital legal advisor; professional, rigorous, prudent; uses legal terms accurately with plain-language explanations.
@@ -122,7 +123,8 @@ ManageAgent({
  "name": "Legal Advisor Assistant",
  "description": "A digital Agent focused on contract review and legal risk assessment",
  "persona": { "L0": "…", "L1": { "role": "…", "personality": ["professional","rigorous","prudent"], "communication_style": "…" } },
-  "principles": { "L0": "…", "L1": { "must_do": ["…"], "must_not": ["…"], "priority": "user safety > accuracy > efficiency" } }
+  "principles": { "L0": "…", "L1": { "must_do": ["…"], "must_not": ["…"], "priority": "user safety > accuracy > efficiency" } },
+  "instructions": "# Responsibilities\nDescribe stable duties here.\n\n## Delivery requirements\nInclude the result location and evidence."
 })
 ```

@@ -189,3 +191,5 @@ Present in a user-friendly way (no internal paths / technical details): confirm

 - AgentFS structure, troubleshooting, and protected paths: see `_agentfs-background.md` and `_protected-paths.yaml`.
 - Always create via `ManageAgent`; never use curl / HTTP or write AgentFS directories directly. Requires client ≥ 10.0.90.
+
+Keep role/expression in persona, boundaries in principles, stable duties in instructions, and methods in skills. Never place private user or team data in distributable Agent files.
--- a/skills/create-agent/SKILL.zh-CN.md
+++ b/skills/create-agent/SKILL.zh-CN.md
@@ -8,7 +8,7 @@

 ## L1：概述

-元技能：多轮对话收集需求 → 生成 persona/principles → `ManageAgent(action='create')` 落地。用于为用户培养专业领域数字助手（法律顾问、财务分析师）、企业快速部署定制业务 Agent、或基于模板快速创建原型。创建的仓库符合 AgentFS v2 规范、由 git 管理版本（可治理、可追溯）。价值在工具给不了的部分：领域化的人格/原则生成 + 创建前的预览确认。
+元技能：多轮对话收集需求 → 生成 persona/principles 与可选长期指令 → `ManageAgent(action='create')` 落地。用于为用户培养专业领域数字助手（法律顾问、财务分析师）、企业快速部署定制业务 Agent、或基于模板快速创建原型。创建的仓库符合 AgentFS v2 规范、由 git 管理版本（可治理、可追溯）。价值在工具给不了的部分：领域化的人格/原则生成 + 创建前的预览确认。

 ## L2：详细规范

@@ -26,9 +26,10 @@

 ### 阶段 3：内容生成

-按结构化字段组织 persona 与 principles（不要输出 raw markdown，按字段整理后再展示）。未收集到的字段留空，由系统补全默认：
+人格与原则按既有结构化字段组织；长期指令使用普通 Markdown 字符串，先以自然语言说明职责和交付要求。未收集到的字段留空，由系统补全默认：

 - **persona**：L0 一句话核心身份；L1 `role`（角色定位）/ `personality`（特征标签数组）/ `communication_style`（沟通风格）；L2 专业领域、价值观、决策偏好（free-form）。
+- **instructions**：可选普通 Markdown 字符串，维护稳定职责、工作策略与交付契约，全文加载，不使用 L0/L1/L2 或自动截断。详细方法、长 SOP、案例与资料放入技能；只知道名称时省略，不虚构职责。
 - **principles**：L0 一句话最高原则；L1 `must_do` / `must_not`（安全红线）/ `priority`（优先级排序）；L2 治理原则、升级规则（free-form）。

 **领域匹配参考**（据 domain 推荐人格与红线）：
@@ -42,7 +43,7 @@

 ### 阶段 4：用户确认

-以自然语言/表格向用户展示预览（名称、描述、人格设定、行为原则；**不展示 raw markdown 源码**），例如：
+以自然语言/表格向用户展示预览（名称、描述、人格设定、行为原则、可选长期指令；**不展示 raw markdown 源码**），例如：

 > 即将创建「法律顾问小助手」——专注合同审查与法律风险评估。
 > **人格**：数字法律顾问；专业、严谨、审慎；准确用法律术语并给通俗解释。
@@ -61,7 +62,8 @@ ManageAgent({
  "name": "法律顾问小助手",
  "description": "专注于合同审查和法律风险评估的数字智能体",
  "persona": { "L0": "…", "L1": { "role": "…", "personality": ["专业","严谨","审慎"], "communication_style": "…" } },
-  "principles": { "L0": "…", "L1": { "must_do": ["…"], "must_not": ["…"], "priority": "用户安全 > 准确性 > 效率" } }
+  "principles": { "L0": "…", "L1": { "must_do": ["…"], "must_not": ["…"], "priority": "用户安全 > 准确性 > 效率" } },
+  "instructions": "# 职责\n写明稳定负责的工作。\n\n## 交付要求\n给出成果位置与必要证据。"
 })
 ```

@@ -128,3 +130,5 @@ ManageAgent({

 - AgentFS 结构、排查要点与受保护路径详见 `_agentfs-background.md` 与 `_protected-paths.yaml`。
 - 一律经 `ManageAgent` 创建，禁止 curl / HTTP 或直接写 AgentFS 目录；需客户端 ≥ 10.0.90。
+
+角色与表达归人格，边界归原则，稳定职责归长期指令，方法归技能。用户或团队私有数据不得写入可分发 Agent 本体。
--- a/skills/create-agent/catalog-metadata.v1.json
+++ b/skills/create-agent/catalog-metadata.v1.json
@@ -11,12 +11,12 @@
      "zh-CN": {
        "name": "创建智能体",
        "summary": "通过自然语言对话收集需求，一键创建专业化数字智能体",
-        "description": "通过多轮对话收集需求，调用 ManageAgent 内置工具创建新的 AgentFS v2 智能体，支持自定义 persona 和 principles。Use when 用户要求创建新智能体、培养某领域助手、或快速基于模板生成可治理 Agent。可选头像生成需配置图像服务，适用供应商条款/额度及独立费用。"
+        "description": "通过多轮对话收集需求，调用 ManageAgent 内置工具创建新的 AgentFS v2 智能体，支持自定义 persona、principles 和可选 instructions。Use when 用户要求创建新智能体、培养某领域助手、或快速基于模板生成可治理 Agent。可选头像生成需配置图像服务，适用供应商条款/额度及独立费用。"
      },
      "en-US": {
        "name": "Create Agent",
        "summary": "Collect requirements through natural-language conversation and create a specialized digital Agent in one step",
-        "description": "Collect requirements through multi-turn conversation and call the ManageAgent builtin tool to create a new AgentFS v2 Agent, with customizable persona and principles. Use when the user asks to create a new Agent, raise a domain assistant, or quickly produce a governable Agent from a template. Optional avatar generation requires a configured image service under provider terms/quota and separate charges."
+        "description": "Collect requirements through multi-turn conversation and call the ManageAgent builtin tool to create a new AgentFS v2 Agent, with customizable persona, principles and optional standing instructions. Use when the user asks to create a new Agent, raise a domain assistant, or quickly produce a governable Agent from a template. Optional avatar generation requires a configured image service under provider terms/quota and separate charges."
      }
    },
    "category": "productivity",
@@ -28,13 +28,13 @@
  },
  "release": {
    "state": "known",
-    "version": "2.7.0",
+    "version": "2.8.0",
    "versionScheme": "semver"
  },
  "timestamps": {
    "catalogUpdatedAt": {
      "state": "known",
-      "value": "2026-10-03",
+      "value": "2026-10-08",
      "precision": "day"
    },
    "releasePublishedAt": {
@@ -65,7 +65,7 @@
    "platforms": {
      "state": "unknown"
    },
-    "requiredClientVersion": "10.0.108"
+    "requiredClientVersion": "10.0.179"
  },
  "spec": {
    "kind": "skill",