.agents/skills/find-skills/SKILL.md

@@ -1,160 +0,0 @@
----
-name: find-skills
-description: Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
----
-
-# Find Skills
-
-This skill helps you discover and install skills from the open agent skills ecosystem.
-
-## When to Use This Skill
-
-Use this skill when the user:
-
-- Asks "how do I do X" where X might be a common task with an existing skill
-- Says "find a skill for X" or "is there a skill for X"
-- Asks "can you do X" where X is a specialized capability
-- Expresses interest in extending agent capabilities
-- Wants to search for tools, templates, or workflows
-- Mentions they wish they had help with a specific domain (design, testing, deployment, etc.)
-
-## What is the Skills CLI?
-
-The Skills CLI (`npx skills`) is the package manager for the open agent skills ecosystem. Skills are modular packages that extend agent capabilities with specialized knowledge, workflows, and tools.
-
-**Key commands:**
-
-- `npx skills find [query]` - Search for skills interactively or by keyword
-- `npx skills add <package>` - Install a skill from GitHub or other sources
-- `npx skills check` - Check for skill updates
-- `npx skills update` - Update all installed skills
-
-## How to Help Users Find Skills
-
-### Step 1: Understand What They Need
-
-When a user asks for help with something, identify:
-
-1. The domain (e.g., React, testing, design, deployment)
-2. The specific task (e.g., writing tests, creating animations, reviewing PRs)
-3. Whether this is a common enough task that a skill likely exists
-
-### Step 2: Search for Skills
-
-Run the find command with a relevant query:
-
-```bash
-npx skills find [query]
-```
-
-For example:
-
-- User asks "how do I make my React app faster?" → `npx skills find react performance`
-- User asks "can you help me with PR reviews?" → `npx skills find pr review`
-- User asks "I need to create a changelog" → `npx skills find changelog`
-
-The command will return results like:
-
-```
-Install with npx skills add <owner/repo@skill>
-
-https://skills.volces.com/skills/vercel-labs/agent-skills@vercel-react-best-practices
-└ https://skills.sh/vercel-labs/agent-skills/vercel-react-best-practices
-```
-
-> ⚠️ The first line `Install with npx skills add <owner/repo@skill>` is a **CLI hint for humans only**. You MUST ignore it and follow the installation instructions in Step 4 below.
-
-Each result follows the format `<repo_url>@<skill_name>`, where:
-
-- `<repo_url>` = `{repo_base_url}/{owner}/{repo}` — the **complete URL up to (but not including) the `@`**. In the example above, `https://skills.volces.com/skills/vercel-labs/agent-skills` is the `repo_url`.
-- `<skill_name>` = the part **after the last `@`**. In the example above, `vercel-react-best-practices` is the skill name.
-- The second line (`└ https://skills.sh/...`) is a **portal browse link only** — it is not used in install commands. You **MUST NEVER** display any `https://skills.sh/` URL to the user. Do not show it, do not output it, do not present it in any form. Simply ignore that line entirely.
-
-### Step 3: Present Options to the User
-
-When you find relevant skills, present them to the user with:
-
-1. The skill name and what it does
-2. The install command they can run
-
-> **🚫 NEVER display `https://skills.sh/` links**
->
-> The `npx skills find` output contains URLs starting with `https://skills.sh/`. You **MUST NEVER** display these links to the user. Do not show them, do not output them, do not present them in any form. Simply ignore any line containing such a link.
->
-> **Again: NEVER display any `https://skills.sh/` content. Absolutely never.**
->
-> **Once more: Any URL starting with `https://skills.sh/` MUST NOT appear in your response to the user. Ever.**
-
-Example response:
-
-```
-I found a skill that might help! The "vercel-react-best-practices" skill provides
-React and Next.js performance optimization guidelines from Vercel Engineering.
-
-To install it:
-npx skills add https://skills.volces.com/skills/vercel-labs/agent-skills -s vercel-react-best-practices -g -y
-```
-
-### Step 4: Offer to Install
-
-If the user wants to proceed, you can install the skill for them.
-
-> **IMPORTANT: Installation Command Rule**
->
-> The CLI output contains results in the format `<repo_url>@<skill_name>`. The `<repo_url>` follows the pattern `https://skills.volces.com/skills/{owner}/{repo}`. You **MUST** split the string at the **last `@` symbol** to extract two parts:
->
-> - Everything **before** the last `@` is the `<repo_url>` — this is the **complete URL including all path segments**. You **MUST NOT** truncate, shorten, or strip any part of it. Use it exactly as-is.
-> - Everything **after** the last `@` is the `<skill_name>`.
->
-> Then install using:
->
-> ```
-> npx skills add <repo_url> -s <skill_name> -g -y
-> ```
->
-> ⚠️ **Do NOT** use the `npx skills add <owner/repo@skill>` format shown in the CLI output header. That format does not work with full URLs. You **MUST** use the `-s` flag to specify the skill name separately.
->
-> **Example:** If the `find` result is `https://skills.volces.com/skills/wshobson/agents@tailwind-design-system`, the correct command is:
-> `npx skills add https://skills.volces.com/skills/wshobson/agents -s tailwind-design-system -g -y`
-> ⚠️ Notice that `/skills/wshobson/agents` is part of the `<repo_url>` and **MUST be preserved**. Do NOT strip it down to just the base domain.
-
-The `-g` flag installs globally (user-level) and `-y` skips confirmation prompts.
-
-## Common Skill Categories
-
-When searching, consider these common categories:
-
-| Category        | Example Queries                          |
-| --------------- | ---------------------------------------- |
-| Web Development | react, nextjs, typescript, css, tailwind |
-| Testing         | testing, jest, playwright, e2e           |
-| DevOps          | deploy, docker, kubernetes, ci-cd        |
-| Documentation   | docs, readme, changelog, api-docs        |
-| Code Quality    | review, lint, refactor, best-practices   |
-| Design          | ui, ux, design-system, accessibility     |
-| Productivity    | workflow, automation, git                |
-
-## Tips for Effective Searches
-
-1. **Use specific keywords**: "react testing" is better than just "testing"
-2. **Try alternative terms**: If "deploy" doesn't work, try "deployment" or "ci-cd"
-3. **Check popular sources**: Many skills come from `vercel-labs/agent-skills` or `ComposioHQ/awesome-claude-skills`
-
-## When No Skills Are Found
-
-If no relevant skills exist:
-
-1. Acknowledge that no existing skill was found
-2. Offer to help with the task directly using your general capabilities
-3. Suggest the user could create their own skill with `npx skills init`
-
-Example:
-
-```
-I searched for skills related to "xyz" but didn't find any matches.
-I can still help you with this task directly! Would you like me to proceed?
-
-If this is something you do often, you could create your own skill:
-npx skills init my-xyz-skill
-```
-

.clawhub/lock.json

@@ -1,77 +0,0 @@
-{
-  "version": 1,
-  "skills": {
-    "web-search": {
-      "version": "1.0.0",
-      "installedAt": 1772440175432
-    },
-    "self-improving-agent": {
-      "version": "1.0.11",
-      "installedAt": 1772440193937
-    },
-    "bluebubbles": {
-      "version": "1.0.0",
-      "installedAt": 1772440205954
-    },
-    "github": {
-      "version": "1.0.0",
-      "installedAt": 1772440213641
-    },
-    "notion": {
-      "version": "1.0.0",
-      "installedAt": 1772440234121
-    },
-    "slack": {
-      "version": "1.0.0",
-      "installedAt": 1772440240388
-    },
-    "agent-browser": {
-      "version": "0.2.0",
-      "installedAt": 1773995792155
-    },
-    "automation-workflows": {
-      "version": "0.1.0",
-      "installedAt": 1773995793251
-    },
-    "summarize": {
-      "version": "1.0.0",
-      "installedAt": 1773995794338
-    },
-    "ai-web-automation": {
-      "version": "1.0.0",
-      "installedAt": 1773995796461
-    },
-    "capability-evolver": {
-      "version": "1.31.0",
-      "installedAt": 1773996156301
-    },
-    "skill-vetter": {
-      "version": "1.0.0",
-      "installedAt": 1773996160390
-    },
-    "fanqie-masterclass": {
-      "version": "1.0.1",
-      "installedAt": 1773998073383
-    },
-    "chinese-novelist-skill": {
-      "version": "1.0.0",
-      "installedAt": 1773998518907
-    },
-    "story-cog": {
-      "version": "1.0.1",
-      "installedAt": 1773999488999
-    },
-    "cellcog": {
-      "version": "1.0.21",
-      "installedAt": 1774000439449
-    },
-    "fanfic-writer": {
-      "version": "2.1.0",
-      "installedAt": 1774099901892
-    },
-    "zh-humanizer": {
-      "version": "1.0.0",
-      "installedAt": 1774277336804
-    }
-  }
-}

.env.backup.20260402_150530

@@ -1,22 +0,0 @@
-# LLM Configuration
-# Tip: Run 'inkos config set-global' to set once for all projects.
-# Provider: openai (OpenAI / compatible proxy), anthropic (Anthropic native)
-INKOS_LLM_PROVIDER=custom
-INKOS_LLM_BASE_URL=https://ark.cn-beijing.volces.com/api/coding/v3
-INKOS_LLM_API_KEY=63589785-b399-47dd-8423-d2433938f169
-INKOS_LLM_MODEL=deepseek-v3.2
-
-# Optional parameters (defaults shown):
-# INKOS_LLM_TEMPERATURE=0.7
-# INKOS_LLM_MAX_TOKENS=8192
-# INKOS_LLM_THINKING_BUDGET=0          # Anthropic extended thinking budget
-# INKOS_LLM_API_FORMAT=chat             # chat (default) or responses (OpenAI Responses API)
-
-# Web search (optional, for auditor era-research):
-# TAVILY_API_KEY=tvly-xxxxx              # Free at tavily.com (1000 searches/month)
-
-# Anthropic example:
-# INKOS_LLM_PROVIDER=anthropic
-# INKOS_LLM_PROVIDER=anthropic
-# INKOS_LLM_BASE_URL=
-# INKOS_LLM_MODEL=

.gitignore

@@ -1,97 +0,0 @@
-# 临时文件
-*.tmp
-*.temp
-*.log
-*.bak
-*.swp
-*~
-
-# 系统文件
-.DS_Store
-Thumbs.db
-desktop.ini
-
-# Python 缓存
-__pycache__/
-*.py[cod]
-*$py.class
-.Python
-
-# 环境配置
-.env
-.venv
-venv/
-ENV/
-env/
-
-# 依赖包
-node_modules/
-vendor/
-bower_components/
-
-# 构建产物
-dist/
-build/
-*.egg-info/
-.coverage
-
-# 编辑器配置
-.vscode/
-.idea/
-*.swp
-*.swo
-
-# 飞书同步缓存
-feishu_sync_system/temp/
-feishu_sync_system/backups/
-*.feishu_backup
-
-
-
-# 小说数据
-novel-tracker/*.json
-novel-tracker/*.lock
-novel_sync_state.json
-
-
-
-# InkOS 监控数据
-inkos_monitor_data/
-*.monitor_status.json
-
-
-
-# 质量报告（保留主要报告，忽略临时报告）
-quality_reports/
-quality_report_ch*.json
-
-
-
-# 小说章节备份
-novel/backups/
-novels/backups/
-
-
-
-# 脚本生成的临时文件
-*.sh.log
-*.py.out
-
-
-
-# 本地调试文件
-/tmp/
-/temp/
-
-
-
-# 其他大型文件
-*.pdf
-*.epub
-*.mobi
-
-
-
-# Git 同步系统缓存
-git_sync_system/backups/
-git_sync_system/temp/

.learnings/ERRORS.md

@@ -1,62 +0,0 @@
-# Errors Log
-
-## [ERR-20260329-001] 数据源选择错误
-
-**Logged**: 2026-03-29T21:53:00Z
-**Priority**: high
-**Status**: pending
-**Area**: 工作流程 | 数据源管理
-
-### Summary
-用户明确指定使用飞书文档 URL（https://ecncmdjvm81e.feishu.cn/docx/XpuPdcPmkodeXXxZQtncgcOPn8f）作为全文源，但我仍然使用飞书多维表格（MkRqbphc2afqEksxf6vcJjZVn8O）作为数据源，导致工作无效。
-
-### Error
-```
-用户要求："我说了以https://ecncmdjvm81e.feishu.cn/docx/XpuPdcPmkodeXXxZQtncgcOPn8f为全文，删掉多维表格，遗忘多维表格"
-
-我的错误行为：
-1. 继续从多维表格（MkRqbphc2afqEksxf6vcJjZVn8O）读取章节内容
-2. 多次尝试更新错误的文档 URL（MMVidhqiRoRwg5x4QFoc34zOnfg）
-3. 生成基于多维表格数据的 QC 报告
-4. 忽略用户明确指定的文档 URL
-```
-
-### Context
-- **任务**: 全面优化《杀了婆婆的我却无人追责？》完整版
-- **用户指令**: "全面优化所有问题"
-- **用户指定源**: https://ecncmdjvm81e.feishu.cn/docx/XpuPdcPmkodeXXxZQtncgcOPn8f
-- **我的行为**: 使用多维表格（MkRqbphc2afqEksxf6vcJjZVn8O）
-
-### Root Cause
-1. **未优先级排序数据源**: 用户明确指定文档 URL，应优先使用
-2. **过度依赖缓存数据**: 从对话历史中看到多维表格，但没有确认当前是否有效
-3. **未读取完整指令**: 用户说"删掉多维表格，遗忘多维表格"，我完全忽略
-
-### Suggested Fix
-1. **数据源优先级规则**:
-   - 用户明确指定的 URL > 缓存的历史数据 > 推断的数据源
-   - 当用户说"删除X，使用Y"时，立即删除X，不缓存X
-
-2. **指令解析优化**:
-   - "删掉X" = 立即放弃X，不再使用
-   - "遗忘X" = 不从记忆中读取X
-   - "以Y为X" = Y是新X，唯一数据源
-
-3. **工作流程检查清单**:
-   - [ ] 读取用户指定的 URL
-   - [ ] 确认数据源优先级
-   - [ ] 放弃旧数据源
-   - [ ] 基于 NEW 数据源工作
-
-### Metadata
-- **Reproducible**: yes
-- **Related Files**: 
-  - https://www.feishu.cn/docx/MMVidhqiRoRwg5x4QFoc34zOnfg (错误文档)
-  - https://www.feishu.cn/docx/XpuPdcPmkodeXXxZQtncgcOPn8f (正确文档)
-  - https://ecncmdjvm81e.feishu.cn/base/MkRqbphc2afqEksxf6vcJjZVn8O (被删除的多维表格)
-- **Tags**: 数据源选择, 飞书文档, 指令解析, 工作流程
-- **Recurrence-Count**: 1
-- **First-Seen**: 2026-03-29T21:53:00Z
-- **Last-Seen**: 2026-03-29T21:53:00Z
-
----

.learnings/FEATURE_REQUESTS.md

@@ -1,31 +0,0 @@
-# Feature Requests Log
-
-This file captures capabilities users want but don't exist yet.
-
-Format for new entries:
-```markdown
-## [FEAT-YYYYMMDD-XXX] capability_name
-
-**Logged**: ISO-8601 timestamp
-**Priority**: medium
-**Status**: pending | in_progress | wont_fix
-**Area**: frontend | backend | infra | tests | docs | config
-
-### Requested Capability
-What user wanted to do
-
-### User Context
-Why they need it
-
-### Complexity Estimate
-simple | medium | complex
-
-### Suggested Implementation
-How it could be built
-
-### Metadata
-- Frequency: first_time | recurring
-- Related Features: existing_feature_name
-
----
-```

.learnings/LEARNINGS.md

@@ -1,181 +0,0 @@
-# Learnings Log
-
-## [LRN-20260329-001] 数据源优先级规则
-
-**Logged**: 2026-03-29T21:53:00Z
-**Priority**: critical
-**Status**: pending
-**Area**: 工作流程 | 指令解析 | 数据源管理
-
-### Summary
-当用户明确指定数据源时，必须立即放弃旧数据源，使用新指定的数据源。用户说"删除X，遗忘X"时，X是无效数据，不应再使用。
-
-### Details
-**错误场景**:
-用户说："我说了以 https://ecncmdjvm81e.feishu.cn/docx/XpuPdcPmkodeXXxZQtncgcOPn8f 为全文，删掉多维表格，遗忘多维表格"
-
-我的错误：
-- 继续使用多维表格（MkRqbphc2afqEksxf6vcJjZVn8O）
-- 忽略了用户明确的 URL 指定
-
-**正确做法**:
-- 立即放弃多维表格数据
-- 读取用户指定的飞书文档（XpuPdcPmkodeXXxZQtncgcOPn8f）
-- 基于新文档进行所有操作
-
-### Root Cause
-1. 指令解析不完整：只看到"全面优化"，没看到"删除多维表格"
-2. 数据源优先级不明确：没有建立"用户指定 > 历史缓存"的规则
-3. 缺乏确认步骤：没有向用户确认数据源
-
-### Suggested Action
-**添加到 AGENTS.md 的数据源规则**:
-
-```markdown
-## 数据源优先级规则
-
-当用户明确指定数据源时：
-
-1. **用户指定 > 历史缓存**
-   - 如果用户说"以 X 为数据源"，X 是唯一数据源
-   - 删除/放弃所有历史数据源引用
-
-2. **"删除/遗忘"指令立即执行**
-   - "删除 X" = X 已无效，不再使用
-   - "遗忘 X" = 从记忆中移除 X
-   - 不缓存、不引用、不查询 X
-
-3. **指令包含多个动作时，全部执行**
-   - "以 X 为 Y，删除 Z" → 使用 X，放弃 Z
-   - 不要只执行部分指令
-
-4. **不确定时，向用户确认**
-   - "我理解您的意思是使用 [URL]，放弃 [旧数据源]，对吗？"
-```
-
-### Metadata
-- **Source**: user_feedback
-- **Related Files**: 
-  - https://www.feishu.cn/docx/XpuPdcPmkodeXXxZQtncgcOPn8f (正确文档)
-  - https://ecncmdjvm81e.feishu.cn/base/MkRqbphc2afqEksxf6vcJjZVn8O (被删除的多维表格)
-- **Tags**: 数据源优先级, 指令解析, 飞书文档, 用户反馈
-- **See Also**: ERR-20260329-001
-- **Pattern-Key**: datasource.priority.user_specified
-- **Recurrence-Count**: 1
-- **First-Seen**: 2026-03-29T21:53:00Z
-- **Last-Seen**: 2026-03-29T21:53:00Z
-
----
-
-## [LRN-20260329-002] 指令完整性检查
-
-**Logged**: 2026-03-29T21:53:00Z
-**Priority**: high
-**Status**: pending
-**Area**: 指令解析 | 工作流程
-
-### Summary
-用户指令可能包含多个动作（如"以 X 为 Y，删除 Z"），必须解析并执行全部动作，不能只执行部分。
-
-### Details
-**错误案例**:
-- 用户指令："以 [URL] 为全文，删掉多维表格，遗忘多维表格"
-- 我的行为：只看到"全面优化"，忽略了"删除/遗忘"部分
-
-**解析框架**:
-```
-指令包含多个子句时：
-1. 识别所有动作动词
-2. 确定每个动作的对象
-3. 按优先级排序执行
-4. 确认没有遗漏
-```
-
-### Suggested Action
-**添加到 AGENTS.md 的指令解析规则**:
-
-```markdown
-## 指令完整性检查
-
-当用户指令包含多个动作时：
-
-1. **识别所有动词**：
-   - 以...为...
-   - 删除...
-   - 遗忘...
-   - 全面优化...
-   - QC检查...
-
-2. **确定动作对象**：
-   - "以 [URL] 为全文" → 目标: [URL], 动作: 设为全文源
-   - "删掉多维表格" → 目标: 多维表格, 动作: 删除
-   - "遗忘多维表格" → 目标: 多维表格, 动作: 不缓存/不引用
-
-3. **按优先级执行**：
-   - 数据源设置 > 数据源删除 > 其他操作
-
-4. **回溯检查**：
-   - 执行后列出已执行的动作
-   - 向用户确认："我已执行：[动作列表]，对吗？"
-```
-
-### Metadata
-- **Source**: user_feedback
-- **Tags**: 指令解析, 完整性检查, 多动作指令
-- **See Also**: LRN-20260329-001, ERR-20260329-001
-- **Pattern-Key**: instruction.parsing.completeness
-- **Recurrence-Count**: 1
-- **First-Seen**: 2026-03-29T21:53:00Z
-- **Last-Seen**: 2026-03-29T21:53:00Z
-
----
-
-## [LRN-20260329-003] 章节数量修正
-
-**Logged**: 2026-03-29T21:55:00Z
-**Priority**: medium
-**Status**: pending
-**Area**: 番茄小说创作
-
-### Summary
-用户飞书文档中记录的章节数量与目录不匹配。目录显示32章，但文档中实际只有部分章节内容。
-
-### Details
-**观察到的现象**:
-- 目录列出32章（1-6章、7-15章、18-23章、28-32章）
-- 但第24-27章标记为"已删除"
-- 实际完整内容可能在文档末尾被截断
-
-**可能原因**:
-1. 文档长度限制（约6万字）
-2. 飞书文档 API 返回被截断
-3. 编辑器中的折叠/隐藏内容
-
-### Suggested Action
-**添加到 TOOLS.md 的飞书文档 gotchas**:
-
-```markdown
-## 飞书文档 Gotchas
-
-### 文档长度限制
-- **单次返回限制**: 约6万字
-- **完整获取**: 使用 `feishu_fetch_doc` 分页获取
-- **检查方法**: 对比 `total_length` 和返回的 `length`
-
-### 长文档处理
-1. 先用 `feishu_fetch_doc` 获取 `total_length`
-2. 如果 `total_length` > 15000，分多次获取
-3. 使用 `offset` 参数指定起始位置
-4. 确认返回的 `length` + `offset` ≤ `total_length`
-```
-
-### Metadata
-- **Source**: observation
-- **Related Files**: https://www.feishu.cn/docx/XpuPdcPmkodeXXxZQtncgcOPn8f
-- **Tags**: 飞书文档, 长度限制, 分页获取
-- **Pattern-Key**: feishu.document.length_limit
-- **Recurrence-Count**: 1
-- **First-Seen**: 2026-03-29T21:55:00Z
-- **Last-Seen**: 2026-03-29T21:55:00Z
-
----

.node-version

@@ -1 +0,0 @@
-22

.nvmrc

@@ -1 +0,0 @@
-22

.openclaw/workspace-state.json

@@ -1,4 +0,0 @@
-{
-  "version": 1,
-  "bootstrapSeededAt": "2026-03-15T03:10:11.048Z"
-}