feat: release 2.0.5

2026-06-02 00:57:41 +08:00 · 2026-04-01 15:01:53 +08:00
parent b058af122c
commit 830b8f2971
15 changed files with 316 additions and 537 deletions
--- a/skills/README.md
+++ b/skills/README.md
@@ -1,124 +1,89 @@
-# Skills Directory
+# Skills

-This directory contains skills for the COW agent system. Skills are markdown files that provide specialized instructions for specific tasks.
+Skills are reusable instruction sets that extend the agent's capabilities. Each skill is a `SKILL.md` file in its own directory, providing specialized knowledge, workflows, and tool integrations for specific tasks.

-## What are Skills?
+## Skill Hub

-Skills are reusable instruction sets that help the agent perform specific tasks more effectively. Each skill:
+Browse, search, and install skills from [Cow Skill Hub](https://skills.cowagent.ai/).

- Provides context-specific guidance
- Documents best practices
- Includes examples and usage patterns
- Can have requirements (binaries, environment variables, etc.)
+Open source: [github.com/zhayujie/cow-skill-hub](https://github.com/zhayujie/cow-skill-hub)
+
+## Install Skills
+
+Install skills from multiple sources via chat (`/skill`) or terminal (`cow skill`):
+
+```bash
+/skill install <name>                   # From Skill Hub
+/skill install <owner>/<repo>           # From GitHub
+/skill install clawhub:<name>           # From ClawHub
+/skill install linkai:<code>            # From LinkAI
+/skill install <url>                    # From URL (zip or SKILL.md)
+```
+
+List all available remote skills:
+
+```bash
+/skill list --remote
+```
+
+## Manage Skills
+
+```bash
+/skill list                  # List installed skills
+/skill info <name>           # View skill details
+/skill enable <name>         # Enable a skill
+/skill disable <name>        # Disable a skill
+/skill uninstall <name>      # Uninstall a skill
+```
+
+> In terminal, replace `/skill` with `cow skill`.

 ## Skill Structure

-Each skill is a markdown file (`SKILL.md`) in its own directory with frontmatter:
+```
+skills/
+  my-skill/
+    SKILL.md          # Required: skill definition
+    scripts/          # Optional: bundled scripts
+    resources/        # Optional: reference files
+```
+
+`SKILL.md` uses YAML frontmatter:

 ```markdown
 ---
-name: skill-name
+name: my-skill
 description: Brief description of what the skill does
-metadata: {"cow":{"emoji":"🎯","requires":{"bins":["tool"]}}}
+metadata: {"cow":{"emoji":"🔧","requires":{"bins":["tool"],"env":["API_KEY"]}}}
 ---

-# Skill Name
+# My Skill

-Detailed instructions and examples...
+Instructions, examples, and usage patterns...
 ```

-## Available Skills
-
- **calculator**: Mathematical calculations and expressions
- **web-search**: Search the web for current information
- **file-operations**: Read, write, and manage files
-
-## Creating Custom Skills
-
-To create a new skill:
-
-1. Create a directory: `skills/my-skill/`
-2. Create `SKILL.md` with frontmatter and content
-3. Restart the agent to load the new skill
-
 ### Frontmatter Fields

- `name`: Skill name (must match directory name)
- `description`: Brief description (required)
- `metadata`: JSON object with additional configuration
-  - `emoji`: Display emoji
-  - `always`: Always include this skill (default: false)
-  - `primaryEnv`: Primary environment variable needed
-  - `os`: Supported operating systems (e.g., ["darwin", "linux"])
-  - `requires`: Requirements object
-    - `bins`: Required binaries
-    - `env`: Required environment variables
-    - `config`: Required config paths
- `disable-model-invocation`: If true, skill won't be shown to model (default: false)
- `user-invocable`: If false, users can't invoke directly (default: true)
+| Field | Description |
+|---|---|
+| `name` | Skill name (must match directory name) |
+| `description` | Brief description (required) |
+| `metadata.cow.emoji` | Display emoji |
+| `metadata.cow.always` | Always include this skill (default: false) |
+| `metadata.cow.requires.bins` | Required binaries |
+| `metadata.cow.requires.env` | Required environment variables |
+| `metadata.cow.requires.config` | Required config paths |
+| `metadata.cow.os` | Supported OS (e.g., `["darwin", "linux"]`) |

-### Example Skill
+## Skill Loading Order

-```markdown
---
-name: my-tool
-description: Use my-tool to process data
-metadata: {"cow":{"emoji":"🔧","requires":{"bins":["my-tool"],"env":["MY_TOOL_API_KEY"]}}}
---
+Skills are loaded from two locations (higher precedence overrides lower):

-# My Tool Skill
+1. **Builtin skills** (lower): `<project_root>/skills/` — shipped with the codebase
+2. **Custom skills** (higher): `~/cow/skills/` — installed via `cow skill install` or skill creator

-Use this skill when you need to process data with my-tool.
+Skills with the same name in the custom directory override builtin ones.

-## Prerequisites
+## Create & Contribute

- Install my-tool: `pip install my-tool`
- Set `MY_TOOL_API_KEY` environment variable
-
-## Usage
-
-\`\`\`python
-# Example usage
-my_tool_command("input data")
-\`\`\`
-```
-
-## Skill Loading
-
-Skills are loaded from multiple locations with precedence:
-
-1. **Workspace skills** (highest): `workspace/skills/` - Project-specific skills
-2. **Managed skills**: `~/.cow/skills/` - User-installed skills
-3. **Bundled skills** (lowest): Built-in skills
-
-Skills with the same name in higher-precedence locations override lower ones.
-
-## Skill Requirements
-
-Skills can specify requirements that determine when they're available:
-
- **OS requirements**: Only load on specific operating systems
- **Binary requirements**: Only load if required binaries are installed
- **Environment variables**: Only load if required env vars are set
- **Config requirements**: Only load if config values are set
-
-## Best Practices
-
-1. **Clear descriptions**: Write clear, concise skill descriptions
-2. **Include examples**: Provide practical usage examples
-3. **Document prerequisites**: List all requirements clearly
-4. **Use appropriate metadata**: Set correct requirements and flags
-5. **Keep skills focused**: Each skill should have a single, clear purpose
-
-## Workspace Skills
-
-You can create workspace-specific skills in your agent's workspace:
-
-```
-workspace/
-  skills/
-    custom-skill/
-      SKILL.md
-```
-
-These skills are only available when working in that specific workspace.
+See the [Skill Creation docs](https://docs.cowagent.ai/skills/create) for details, or submit your skill to [Skill Hub](https://skills.cowagent.ai/submit).
--- a/skills/linkai-agent/README.md
+++ b/skills/linkai-agent/README.md
@@ -1,258 +0,0 @@
-# LinkAI Agent Skill
-
-这个 skill 允许你调用 LinkAI 平台上的多个应用(App)和工作流(Workflow)，通过简单的配置即可集成多个智能体能力。
-
-## 特性
-
- ✅ **多应用支持** - 在一个配置文件中管理多个 LinkAI 应用/工作流
- ✅ **动态加载** - skill 系统加载时自动从 `config.json` 读取应用列表
- ✅ **自动技能描述** - 所有配置的应用会自动添加到技能描述中
- ✅ **模型切换** - 可以为每个请求指定不同的模型
- ✅ **知识库集成** - 支持应用绑定的知识库
- ✅ **插件能力** - 支持应用启用的各类插件
- ✅ **工作流执行** - 支持执行复杂的多步骤工作流
-
-## 快速开始
-
-### 1. 配置 API Key
-
-```bash
-env_config(action="set", key="LINKAI_API_KEY", value="your-linkai-api-key")
-```
-
-获取 API Key: https://link-ai.tech/console/interface
-
-### 2. 配置应用列表
-
-将 `config.json.template` 复制为 `config.json`：
-
-```bash
-cp config.json.template config.json
-```
-
-编辑 `config.json`，添加你的应用/工作流：
-
-```json
-{
-  "apps": [
-    {
-      "app_code": "G7z6vKwp",
-      "app_name": "通用助手",
-      "app_description": "通用AI助手，可以回答各类问题"
-    },
-    {
-      "app_code": "your_kb_app",
-      "app_name": "产品文档助手",
-      "app_description": "基于产品文档知识库的问答助手"
-    },
-    {
-      "app_code": "your_workflow",
-      "app_name": "数据分析工作流",
-      "app_description": "执行数据清洗、分析和可视化的完整工作流"
-    }
-  ]
-}
-```
-
-**注意：** 修改 `config.json` 后，Agent 在下次加载技能时会自动读取新配置。
-
-### 3. 调用应用
-
-```bash
-bash(command='curl -sS --max-time 120 -X POST "https://api.link-ai.tech/v1/chat/completions" -H "Content-Type: application/json" -H "Authorization: Bearer $LINKAI_API_KEY" -d "{\"app_code\":\"G7z6vKwp\",\"messages\":[{\"role\":\"user\",\"content\":\"What is artificial intelligence?\"}],\"stream\":false}"', timeout=130)
-```
-
-## 使用示例
-
-### 基础调用
-
-```bash
-# 调用默认模型 (通过 bash + curl)
-bash(command='curl -sS --max-time 120 -X POST "https://api.link-ai.tech/v1/chat/completions" -H "Content-Type: application/json" -H "Authorization: Bearer $LINKAI_API_KEY" -d "{\"app_code\":\"G7z6vKwp\",\"messages\":[{\"role\":\"user\",\"content\":\"解释一下量子计算\"}],\"stream\":false}"', timeout=130)
-```
-
-### 指定模型
-
-在 JSON body 中添加 `model` 字段：
-
-```json
-{
-  "app_code": "G7z6vKwp",
-  "model": "LinkAI-4.1",
-  "messages": [{"role": "user", "content": "写一篇关于AI的文章"}],
-  "stream": false
-}
-```
-
-### 调用工作流
-
-工作流的 app_code 从 LinkAI 控制台获取，调用方式与普通应用相同。
-
-## ⚠️ 重要提示
-
-### 超时配置
-
-LinkAI 应用（特别是视频/图片生成、复杂工作流）可能需要较长时间处理。在 curl 命令中加入 `--max-time 180`，并相应增加 bash 工具的 `timeout` 参数。
-
-## 配置说明
-
-### config.json 字段
-
-| 字段 | 类型 | 说明 |
-|------|------|------|
-| `app_code` | string | 应用或工作流的唯一标识码，从 LinkAI 控制台获取 |
-| `app_name` | string | 应用名称，会显示在技能描述中 |
-| `app_description` | string | 应用功能描述，帮助 Agent 理解何时使用该应用 |
-
-### 获取 app_code
-
-1. 登录 [LinkAI 控制台](https://link-ai.tech/console)
-2. 进入「应用管理」或「工作流管理」
-3. 选择要集成的应用/工作流
-4. 在应用详情页找到 `app_code`
-
-## 应用类型
-
-### 1. 普通应用
-
-配置了系统提示词和参数的标准对话应用，可以：
- 设置角色和性格
- 绑定知识库
- 启用插件（图像识别、网页搜索、代码执行等）
-
-### 2. 知识库应用
-
-基于特定知识库的问答应用，适合：
- 企业内部知识库
- 产品文档问答
- 客户支持
-
-### 3. 工作流
-
-多步骤的自动化流程，可以：
- 串联多个处理节点
- 条件分支
- 循环处理
- 调用外部 API
-
-## 响应格式
-
-### 成功响应
-
-API 返回 OpenAI 兼容格式，从 `choices[0].message.content` 获取回复内容：
-
-```json
-{
-  "choices": [{
-    "message": {
-      "role": "assistant",
-      "content": "人工智能（AI）是计算机科学的一个分支..."
-    }
-  }],
-  "usage": {
-    "prompt_tokens": 10,
-    "completion_tokens": 150,
-    "total_tokens": 160
-  }
-}
-```
-
-### 错误响应
-
-```json
-{
-  "error": {
-    "message": "应用不存在",
-    "code": "xxx"
-  }
-}
-```
-
-## 常见错误
-
-### LINKAI_API_KEY environment variable is not set
-**原因：** 未配置 API Key  
-**解决：** 使用 `env_config` 工具设置 LINKAI_API_KEY
-
-### 应用不存在 (402)
-**原因：** app_code 不正确或应用已删除  
-**解决：** 检查 app_code 是否正确，确认应用存在
-
-### 无访问权限 (403)
-**原因：** 尝试访问他人的私有应用  
-**解决：** 确保应用是公开的或你是创建者
-
-### 账号积分额度不足 (406)
-**原因：** LinkAI 账户余额不足  
-**解决：** 前往控制台充值
-
-### 内容审核不通过 (409)
-**原因：** 请求或响应包含敏感内容  
-**解决：** 修改输入内容，避免敏感词
-
-## 技术实现
-
-### 自动技能描述生成
-
-当 skill 系统加载 `linkai-agent` 时，会自动：
-1. 读取 `config.json` 中的应用列表
-2. 将每个应用的 name 和 description 动态添加到技能描述中
-3. Agent 加载时会看到完整的应用列表
-
-这是在 `agent/skills/loader.py` 中实现的特殊处理。
-
-### 工作流程
-
-```
-用户配置 config.json
-  ↓
-Agent 启动/重新加载技能
-  ↓
-SkillLoader 检测到 linkai-agent
-  ↓
-动态读取 config.json
-  ↓
-生成包含所有应用描述的 description
-  ↓
-Agent 看到所有可用应用的完整信息
-  ↓
-用户请求触发
-  ↓
-Agent 根据描述选择合适的应用
-  ↓
-通过 bash + curl 调用 LinkAI API
-  ↓
-LinkAI API 处理并返回结果
-```
-
-## 最佳实践
-
-1. **清晰的描述** - 为每个应用写清晰、具体的描述，帮助 Agent 理解应用用途
-2. **合理分工** - 不同应用负责不同领域，避免功能重叠
-3. **无需重启** - 修改 config.json 后，Agent 下次加载技能时会自动更新
-4. **模型选择** - 根据任务复杂度选择合适的模型
-5. **知识库优化** - 为专业领域的应用绑定相关知识库
-
-## 扩展用法
-
-### 在 Agent 系统中使用
-
-当 Agent 系统加载这个 skill 时，会自动从 `config.json` 读取应用列表并生成描述：
-
-```
-Call LinkAI apps/workflows. 通用助手(G7z6vKwp: 通用AI助手，可以回答各类问题); 产品文档助手(kb_app_001: 基于产品文档知识库的问答助手); 数据分析工作流(wf_002: 执行数据清洗、分析和可视化的完整工作流)
-```
-
-Agent 会根据用户问题自动选择最合适的应用进行调用。
-
-## 相关链接
-
- LinkAI 平台: https://link-ai.tech
- API 文档: https://docs.link-ai.tech
- 控制台: https://link-ai.tech/console
- 模型列表: https://link-ai.tech/console/models
- 应用广场: https://link-ai.tech/square
-
-## License
-
-Part of the chatgpt-on-wechat project.
--- a/skills/linkai-agent/SKILL.md
+++ b/skills/linkai-agent/SKILL.md
@@ -1,85 +0,0 @@
---
-name: linkai-agent
-description: Call LinkAI applications and workflows. Use bash with curl to invoke the chat completions API.
-homepage: https://link-ai.tech
-metadata:
-  emoji: 🤖
-  default_enabled: false
-  requires:
-    bins: ["curl"]
-    env: ["LINKAI_API_KEY"]
---
-
-# LinkAI Agent
-
-Call LinkAI applications and workflows through the chat completions API. Available apps are loaded from config.json.
-
-## Setup
-
-This skill requires a LinkAI API key.
-
-1. Get your API key from [LinkAI Console](https://link-ai.tech/console/interface)
-2. Set the environment variable: `export LINKAI_API_KEY=Link_xxxxxxxxxxxx` (or use env_config tool)
-
-## Configuration
-
-1. Copy `config.json.template` to `config.json`
-2. Add your apps/workflows in config.json. The skill description is auto-generated from this config when loaded.
-
-## Usage
-
-Use the bash tool with curl to call the API. **Prefer curl** to avoid encoding issues on Windows PowerShell.
-
-```bash
-curl -X POST "https://api.link-ai.tech/v1/chat/completions" \
-  -H "Content-Type: application/json" \
-  -H "Authorization: Bearer $LINKAI_API_KEY" \
-  -d '{
-    "app_code": "<app_code>",
-    "messages": [{"role": "user", "content": "<question>"}],
-    "stream": false
-  }'
-```
-
-**Optional parameters**:
-
- Add `--max-time 120` to curl for long-running tasks (video/image generation)
-
-**On Windows cmd**: Use `%LINKAI_API_KEY%` instead of `$LINKAI_API_KEY`.
-
-**Example** (via bash tool):
-
-```bash
-bash(command='curl -sS --max-time 120 -X POST "https://api.link-ai.tech/v1/chat/completions" -H "Content-Type: application/json" -H "Authorization: Bearer $LINKAI_API_KEY" -d "{\"app_code\":\"G7z6vKwp\",\"messages\":[{\"role\":\"user\",\"content\":\"What is AI?\"}],\"stream\":false}"', timeout=130)
-```
-
-## Response
-
-Success (extract `choices[0].message.content` from JSON):
-
-```json
-{
-  "choices": [{
-    "message": {
-      "role": "assistant",
-      "content": "AI stands for Artificial Intelligence..."
-    }
-  }],
-  "usage": {
-    "prompt_tokens": 10,
-    "completion_tokens": 50,
-    "total_tokens": 60
-  }
-}
-```
-
-Error:
-
-```json
-{
-  "error": {
-    "message": "Error description",
-    "code": "error_code"
-  }
-}
-```
--- a/skills/linkai-agent/config.json.template
+++ b/skills/linkai-agent/config.json.template
@@ -1,14 +0,0 @@
-{
-  "apps": [
-    {
-      "app_code": "G7z6vKwp",
-      "app_name": "LinkAI客服助手",
-      "app_description": "当用户需要了解LinkAI平台相关问题时才选择该助手，基于LinkAI知识库进行回答"
-    },
-    {
-      "app_code": "SFY5x7JR",
-      "app_name": "内容创作助手",
-      "app_description": "当用户需要创作图片或视频时才使用该助手，支持Nano Banana、Seedream、即梦、Veo、可灵等多种模型"
-    }
-  ]
-}