dijk-skills 1.0.3 → 1.0.5

package/README.md

@@ -19,23 +19,31 @@ npx dijk-skills install [options]
 
 | Option | Description | Default |
 |--------|-------------|---------|
-| `--modules <list>` | Comma-separated module names | `common` |
+| `--modules <list>` | Comma-separated module names | `all` |
 | `--dir <path>` | Target installation directory | `~/.agents` |
 
 ### Modules
 
-| Module | Skills |
-|--------|--------|
-| `common` | grilling, humanizer-zh, i-have-adhd |
-| `writing` | grill-my-proposal, vibe-opinion-insight |
-| `tools` | darwin-skill, skill-creator |
-| `coding` | *(empty, reserved for future skills)* |
+| Module | Skill | Description |
+|--------|-------|-------------|
+| `common` | caveman | 超压缩交流模式,在保持技术准确性的同时大幅节省输出 token(分 lite/full/ultra 档) |
+| `common` | grilling | 对计划、决策或想法进行不留情面的连环追问,压力测试你的思路 |
+| `common` | humanizer-zh | 去除文本中的 AI 生成痕迹,使其更自然、更像人类书写 |
+| `common` | i-have-adhd | 面向 ADHD 读者塑造输出:下一步行动置顶、多步任务编号、复述状态、抑制离题、给出时间预估 |
+| `writing` | grill-my-proposal | 把方案往死里问——三阶段递进追问(宏观结构→逐章深挖→交叉验证),动手前压力测试方案 |
+| `writing` | last30days | 研究任意话题近 30 天的真实讨论,覆盖 Reddit、X、YouTube、TikTok、Hacker News、Polymarket、GitHub 与全网 |
+| `writing` | vibe-code-insight | 对软件项目做深度系统分析,产出 `design.md`(系统设计)与 `trend.md`(生态与趋势洞察) |
+| `writing` | vibe-opinion-insight | Human 驱动的 AI Agent 技术洞察框架——Agent 是启发工具,Human 赋予意义并承担责任 |
+| `tools` | handoff | 将当前会话压缩为交接文档,供另一个 agent 接续工作 |
+| `tools` | mcp-builder | 构建高质量 MCP 服务器的指南,支持 Python(FastMCP)与 Node/TypeScript(MCP SDK) |
+| `tools` | skill-creator | 创建、编辑、评测与优化 skill,含触发准确率的基准测试 |
+| `coding` | *(reserved)* | 空——保留给未来的 skill |
 
 > Installing any module automatically includes `common`.
 
 ### Examples
 
-Install common skills to default location:
+Install all skills to default location:
 ```bash
 npx dijk-skills install
 ```
@@ -57,9 +65,10 @@ Skills are installed to `{dir}/skills/{skill-name}/`:
 ```
 ~/.agents/
 └── skills/
+    ├── caveman/
     ├── grilling/
+    ├── handoff/
     ├── humanizer-zh/
     ├── i-have-adhd/
-    ├── darwin-skill/
     └── ...
 ```

package/bin/dijk-skills.js

@@ -13,7 +13,7 @@ const { values, positionals } = parseArgs({
   options: {
     modules: {
       type: 'string',
-      default: 'common'
+      default: 'all'
     },
     dir: {
       type: 'string',
@@ -37,7 +37,7 @@ Commands:
 Options:
   --modules <list>    Comma-separated module names or "all"
                       Values: coding,common,tools,writing,all
-                      Default: common
+                      Default: all
   --dir <path>        Target installation directory
                       Default: ~/.agents
   -h, --help          Show this help message

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "dijk-skills",
-  "version": "1.0.3",
+  "version": "1.0.5",
   "description": "CLI installer for Dijk skills collection",
   "type": "module",
   "bin": {

package/tools/skills/handoff/SKILL.md

@@ -0,0 +1,16 @@
+---
+name: handoff
+description: Compact the current conversation into a handoff document for another agent to pick up.
+argument-hint: "What will the next session be used for?"
+disable-model-invocation: true
+---
+
+Write a handoff document summarising the current conversation so a fresh agent can continue the work. Save to the temporary directory of the user's OS - not the current workspace.
+
+Include a "suggested skills" section in the document, naming which skills the next agent should call the Skill tool for.
+
+Do not duplicate content already captured in other artifacts (specs, plans, ADRs, issues, commits, diffs). Reference them by path or URL instead.
+
+Redact any sensitive information, such as API keys, passwords, or personally identifiable information.
+
+If the user passed arguments, treat them as a description of what the next session will focus on and tailor the doc accordingly.

package/tools/skills/handoff/agents/openai.yaml

@@ -0,0 +1,5 @@
+interface:
+  display_name: "Handoff"
+  short_description: "Compact a conversation into a handoff"
+policy:
+  allow_implicit_invocation: false