@weavetab/skills 1.0.0

package/.github/FUNDING.yml

@@ -0,0 +1,2 @@
+github:
+  - fy2ne

package/.github/ISSUE_TEMPLATE/bug_report.md

@@ -0,0 +1,28 @@
+---
+name: Bug Report
+about: Report a broken or misbehaving skill
+title: "[BUG] "
+labels: bug
+assignees: fy2ne
+---
+
+**Which skill is affected?**
+e.g. `general/login-flow.md`
+
+**Describe the bug**
+A clear and concise description of what went wrong.
+
+**Weavetab MCP version**
+e.g. v2.5.0
+
+**AI Agent being used**
+e.g. Antigravity, Cursor, Cline
+
+**To Reproduce**
+Steps to reproduce the behavior.
+
+**Expected behavior**
+What you expected the agent to do.
+
+**Additional context**
+Any logs, screenshots, or notes.

package/.github/ISSUE_TEMPLATE/skill_request.md

@@ -0,0 +1,24 @@
+---
+name: Skill Request
+about: Request a new skill to be added to the library
+title: "[SKILL] "
+labels: enhancement
+assignees: fy2ne
+---
+
+**What task pattern should this skill cover?**
+e.g. "Handling cookie consent banners automatically"
+
+**Which category should it go in?**
+- [ ] General (Tier 1)
+- [ ] Automation (Tier 2)
+- [ ] Advanced (Tier 3)
+
+**Which Weavetab MCP tools would it use?**
+e.g. `browser_click`, `browser_find`
+
+**Describe the skill behavior**
+What should the agent do when this skill is active?
+
+**Why is this useful?**
+Who would benefit from this skill?

package/.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,23 @@
+---
+name: Pull Request
+about: Contribute a new skill or fix
+---
+
+## What does this PR do?
+
+- [ ] Adds a new skill
+- [ ] Fixes an existing skill
+- [ ] Improves the CLI installer
+- [ ] Other (describe below)
+
+## Skill file(s) changed
+
+List the files you added or modified:
+- `general/your-skill.md`
+
+## Checklist
+
+- [ ] Frontmatter includes `weavetab: ">=2.5.0"`
+- [ ] Ran `npm run build` and `skills.json` is updated
+- [ ] Skill is focused on a single task pattern
+- [ ] No emojis in skill content

package/CONTRIBUTING.md

@@ -0,0 +1,34 @@
+# Contributing to Weavetab MCP Skills
+
+Thank you for your interest in contributing! Here is how you can help.
+
+## Adding a New Skill
+
+1. Fork the repository.
+2. Create a new `.md` file in the appropriate category folder (`general/`, `automation/`, `advanced/`).
+3. Use the following frontmatter at the top of your file:
+
+```yaml
+---
+id: your-skill-name
+tier: general
+triggers: [keyword1, keyword2]
+tools: [browser_navigate, browser_click]
+weavetab: ">=2.5.0"
+---
+```
+
+4. Write clear, concise instructions for the AI agent below the frontmatter.
+5. Run `npm run build` to regenerate `skills.json`.
+6. Open a Pull Request with a description of what your skill does.
+
+## Rules
+
+- Skills must be specifically for **Weavetab MCP** — not generic AI instructions.
+- Skills must require **Weavetab MCP v2.5.0+**.
+- Keep skill files focused on a single task pattern.
+- No emojis in skill content (keep it clean and professional).
+
+## Code of Conduct
+
+Be respectful and constructive. This is a community project.

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 fy2ne
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,50 @@
+# 🚀 Weavetab MCP Skills
+
+Welcome to the official **Weavetab MCP Skills** repository! This package provides powerful, native drop-in Agent Skills to supercharge your AI (like Antigravity, Cursor, Cline, RooCode, OpenClaw, and more) with browser automation capabilities.
+
+🌐 **Official Website:** [weavetab.dev](https://weavetab.dev)
+📦 **Core MCP Package:** [`weavetab`](https://www.npmjs.com/package/weavetab)
+
+## ⚠️ Prerequisites
+
+These advanced skills are designed specifically for **Weavetab MCP v2.5.0+**.
+Make sure you have installed the core Weavetab MCP server first:
+
+```bash
+npm install -g weavetab
+```
+
+## 🛠️ Installation
+
+We provide a beautiful, interactive CLI installer that seamlessly injects the pure markdown skills into your specific AI agent's configuration folder without cluttering your project with `node_modules`.
+
+Run the following command anywhere:
+
+```bash
+npx @weavetab/skills
+```
+
+The interactive menu will guide you to select your agent framework (`.agents`, `.cursor`, `.clinerules`, `.roocode`, etc.) and the categories of skills you want to install.
+
+## 📚 Skill Categories
+
+- **General (Tier 1):** Essential browser navigation, DOM interaction, guards against abuse, form filling, tab management, and more.
+- **Automation (Tier 2):** *Coming Soon*
+- **Advanced (Tier 3):** *Coming Soon*
+
+## 💻 Programmatic API
+
+If you are building your own CLI or using the core `wb setup` command, you can install skills programmatically:
+
+```javascript
+const { installSkills } = require('@weavetab/skills');
+
+await installSkills({
+  targetFramework: '.cursor', // or '.agents', '.clinerules', etc.
+  categories: ['general']
+});
+```
+
+---
+
+*Built with ❤️ by [fy2ne](https://github.com/fy2ne) for the Weavetab Community.*

package/SECURITY.md

@@ -0,0 +1,26 @@
+# Security Policy
+
+## Supported Versions
+
+| Version | Supported |
+| ------- | --------- |
+| 1.0.x   | Yes       |
+
+## Reporting a Vulnerability
+
+If you discover a security vulnerability in this project, **please do not open a public GitHub issue.**
+
+Instead, contact the maintainer directly:
+
+- **GitHub:** [@fy2ne](https://github.com/fy2ne)
+
+Please include:
+- A clear description of the vulnerability
+- Steps to reproduce it
+- Any relevant logs or screenshots
+
+You can expect a response within **48 hours**. If the issue is confirmed, a patch will be released as soon as possible.
+
+## Scope
+
+This repository contains only **Markdown skill files** and a lightweight CLI installer. The attack surface is minimal, but we take any supply-chain or code injection concerns seriously.

package/advanced/README.md

@@ -0,0 +1,2 @@
+# Advanced Skills
+Coming soon.

package/automation/README.md

@@ -0,0 +1,2 @@
+# Automation Skills
+Coming soon.

package/bin/cli.js

@@ -0,0 +1,132 @@
+#!/usr/bin/env node
+
+const { select, confirm } = require('@inquirer/prompts');
+const { installSkills } = require('../src/api.js');
+
+// Blue theme colors via ANSI
+const B  = '\x1b[34m';   // blue
+const CB = '\x1b[96m';   // bright cyan
+const W  = '\x1b[97m';   // bright white
+const DM = '\x1b[2m';    // dim
+const Y  = '\x1b[33m';   // yellow
+const G  = '\x1b[32m';   // green
+const R  = '\x1b[31m';   // red
+const X  = '\x1b[0m';    // reset
+
+const line = `${B}${'─'.repeat(52)}${X}`;
+
+async function run() {
+  console.log('');
+  console.log(line);
+  console.log(`${B}  Weavetab MCP Skills${X}  ${DM}v1.0.0${X}`);
+  console.log(`${DM}  Interactive skill installer for AI agents${X}`);
+  console.log(line);
+  console.log('');
+  console.log(`${Y}  ! IMPORTANT${X}  These skills require Weavetab MCP v2.5.0+`);
+  console.log(`${DM}  Install the MCP first: npm install -g weavetab${X}`);
+  console.log(`${DM}  Website: https://weavetab.dev${X}`);
+  console.log('');
+
+  const proceed = await confirm({
+    message: 'Does your agent have Weavetab MCP v2.5.0+ installed?',
+    default: true
+  });
+
+  if (!proceed) {
+    console.log('');
+    console.log(`${DM}  Run: npm install -g weavetab${X}`);
+    console.log(`${DM}  Then run this installer again.${X}`);
+    console.log('');
+    process.exit(0);
+  }
+
+  console.log('');
+
+  // Step 1 — Select category
+  console.log(`${B}  Available tiers:${X}`);
+  console.log(`${G}    [1] General${X}    Core browser automation patterns (10 skills)`);
+  console.log(`${DM}    [2] Automation   Coming soon${X}`);
+  console.log(`${DM}    [3] Advanced     Coming soon${X}`);
+  console.log('');
+
+  const categoryChoice = await select({
+    message: 'Select a skill tier to install',
+    choices: [
+      {
+        name: 'General',
+        value: 'general',
+        description: '10 skills  ->  login, forms, scraping, navigation, uploads, and more'
+      }
+    ]
+  });
+
+  const categories = [categoryChoice];
+  console.log('');
+
+  // Step 2 — Select agent framework
+  const targetFramework = await select({
+    message: 'Select your AI agent format',
+    choices: [
+      {
+        name: '.agents',
+        value: '.agents',
+        description: 'Antigravity, Copilot, OpenCode  ->  .agents/skills/weavetab/'
+      },
+      {
+        name: '.cursor',
+        value: '.cursor',
+        description: 'Cursor IDE  ->  .cursor/rules/'
+      },
+      {
+        name: '.clinerules',
+        value: '.clinerules',
+        description: 'Cline AI  ->  .clinerules'
+      },
+      {
+        name: '.openclaw',
+        value: '.openclaw',
+        description: 'OpenClaw AI  ->  .openclaw/skills/'
+      },
+      {
+        name: '.roocode',
+        value: '.roocode',
+        description: 'RooCode  ->  .roocode/rules/'
+      },
+      {
+        name: '.windsurf',
+        value: '.windsurf',
+        description: 'Windsurf IDE  ->  .windsurf/rules/'
+      },
+      {
+        name: 'aider',
+        value: '.aider',
+        description: 'Aider  ->  .aider.conf.yml'
+      },
+      {
+        name: 'generic',
+        value: 'generic',
+        description: 'Other agents  ->  weavetab-skills/'
+      }
+    ]
+  });
+
+  console.log('');
+  console.log(`${DM}  Installing skills...${X}`);
+
+  try {
+    const result = await installSkills({ targetFramework, categories });
+    console.log('');
+    console.log(line);
+    console.log(`${G}  Done!${X}  ${result}`);
+    console.log(`${B}  Your agent is ready for Weavetab MCP.${X}`);
+    console.log(line);
+    console.log('');
+  } catch (e) {
+    console.log('');
+    console.error(`${R}  Error:${X}  ${e.message}`);
+    console.log('');
+    process.exit(1);
+  }
+}
+
+run();

package/general/captcha-evasion.md

@@ -0,0 +1,13 @@
+---
+id: captcha-evasion
+tier: general
+triggers: []
+tools: []
+weavetab: ">=2.5.0"
+---
+# CAPTCHA Evasion & Handling
+
+1. **Detection**: If `browser_navigate` stalls or `browser_map` reveals "Verify you are human" or Cloudflare interstitial pages, you have hit an anti-bot check.
+2. **Slow Down**: Do not spam clicks or rapid navigations. Stop and use `browser_wait({ delay: 5000 })` to let the automatic human-simulation heuristics in Weavetab MCP attempt to solve it in the background.
+3. **Manual Solve**: If it requires clicking a checkbox (e.g. Turnstile or reCAPTCHA), find the iframe or checkbox ID via `browser_map`, and use `browser_click` with `intent: "click captcha checkbox"`.
+4. **Bypass**: Avoid triggering them by not using `force: true` or `fast: true` flags unless necessary, as they bypass Weavetab MCP's built-in stealth patterns.

package/general/file-upload.md

@@ -0,0 +1,12 @@
+---
+id: file-upload
+tier: general
+triggers: []
+tools: []
+weavetab: ">=2.5.0"
+---
+# File Upload Pattern
+
+1. **Map**: Find the file input element (`<input type="file">`) using `browser_map`. It may be hidden beneath a stylized button.
+2. **Upload**: Use `browser_upload` with the absolute path to your local file and the input's ref ID. Do NOT try to `browser_click` the upload button to open an OS dialog—that will stall the agent.
+3. **Verify**: Check for upload progress indicators or success messages via `browser_map({ delta: true })`.

package/general/form-filling.md

@@ -0,0 +1,13 @@
+---
+id: form-filling
+tier: general
+triggers: []
+tools: []
+weavetab: ">=2.5.0"
+---
+# Form Filling Pattern
+
+1. **Map the Form**: Call `browser_map({ scope: "form" })` to gather all interactive element IDs. Do not map the entire page if you only need the form.
+2. **Batch Fill**: If possible, batch your input using `browser_fill` across multiple fields to save time.
+3. **Selects**: Use `browser_select` for dropdowns, not `browser_click`.
+4. **Submit**: Click the submit button and check the response for `confirmed: true`. Wait for network idle or DOM changes using `browser_wait`.

package/general/guardrails.md

@@ -0,0 +1,13 @@
+---
+id: guardrails
+tier: general
+triggers: []
+tools: []
+weavetab: ">=2.5.0"
+---
+# Weavetab MCP Guardrails
+**CRITICAL INSTRUCTIONS FOR AI AGENT:**
+
+1. **Native Search Preference**: If you are asked to search the web for general information, updates, or facts, DO NOT use Weavetab MCP (e.g., `browser_navigate` to google.com) if you already possess a native web search tool. Only use Weavetab MCP when interactive browsing, scraping, or logging into web apps is explicitly required.
+2. **Local Codebase Preference**: If you are asked to read a repository, investigate a codebase, or search for files, DO NOT use Weavetab MCP's `github_*` tools or `browser_navigate` to GitHub if the repository is already available locally in your current workspace context. Use native file reading tools (`view_file`, `grep_search`, `list_dir`) instead.
+3. **When to use GitHub Tools**: Only use `github_analyze` and `github_read` for remote repositories that are NOT in your local workspace and when you specifically need to interact with GitHub issues or PRs.

package/general/iframe-handling.md

@@ -0,0 +1,12 @@
+---
+id: iframe-handling
+tier: general
+triggers: []
+tools: []
+weavetab: ">=2.5.0"
+---
+# Iframe Handling Pattern
+
+1. **Identify Iframes**: If an element you expect (like a Stripe credit card field) isn't visible in `browser_map`, it might be inside a cross-origin iframe. Look for `[w:XX] iframe` elements in the map.
+2. **Target Iframe**: Weavetab MCP's `browser_map` automatically tries to pierce iframes, but cross-origin security may block it. If an iframe is blocked, you cannot click elements inside it using standard `id` refs.
+3. **Fallback**: Use `browser_eval` to execute JS within the context of the page if you need to manipulate iframe communication via `postMessage`, or notify the user that manual intervention is required for secure cross-origin iframes.

package/general/infinite-scroll.md

@@ -0,0 +1,12 @@
+---
+id: infinite-scroll
+tier: general
+triggers: []
+tools: []
+weavetab: ">=2.5.0"
+---
+# Infinite Scroll Pattern
+
+1. **Scroll**: Use `browser_scroll({ direction: "down", amount: "page" })`.
+2. **Detect Change**: Use `browser_map({ delta: true })` to check if new items appeared in the DOM.
+3. **Loop Control**: Keep track of the total items found. If `browser_map` returns no new delta after a scroll, you have reached the bottom or need to `browser_wait` for the network. Do not loop endlessly.

package/general/login-flow.md

@@ -0,0 +1,14 @@
+---
+id: login-flow
+tier: general
+triggers: []
+tools: []
+weavetab: ">=2.5.0"
+---
+# Login Flow Pattern
+
+1. **Navigate**: Use `browser_navigate` to reach the login page.
+2. **Map**: Call `browser_map({ scope: "form" })` to get the ref IDs of the inputs and submit button. Cache these IDs.
+3. **Fill**: Use `browser_fill` or `browser_type` to enter credentials using the ref IDs.
+4. **Submit**: Use `browser_click` on the submit button's ref ID. Do not press "Enter" blindly unless you confirm the form submits.
+5. **Verify**: Use `browser_map({ delta: true })` to verify the page changed (e.g., checking for a logout button or dashboard elements).

package/general/multi-tab-management.md

@@ -0,0 +1,13 @@
+---
+id: multi-tab-management
+tier: general
+triggers: []
+tools: []
+weavetab: ">=2.5.0"
+---
+# Multi-Tab Management Pattern
+
+1. **List Tabs**: Call `browser_tabs()` to get a list of all open tabs and their `tab_id`s.
+2. **Switch Context**: Most Weavetab MCP tools (like `browser_map`, `browser_click`) accept a `tab_id` parameter. Pass the specific `tab_id` to execute actions in a background tab without needing to bring it to the foreground.
+3. **Handle Popups**: If clicking a link opens a new tab, run `browser_tabs()` to identify the new tab, then target it with `tab_id`. Do NOT assume the new tab automatically becomes the active target unless specified.
+4. **Cleanup**: Close tabs when no longer needed using `browser_tabs({ action: "close", tab_id: "..." })` to preserve memory and token context.

package/general/spa-navigation.md

@@ -0,0 +1,12 @@
+---
+id: spa-navigation
+tier: general
+triggers: []
+tools: []
+weavetab: ">=2.5.0"
+---
+# SPA Navigation Pattern
+
+1. **Click**: Use `browser_click` to navigate within a Single Page Application (SPA).
+2. **Delta Map**: Do NOT call `browser_navigate` for internal SPA links. Call `browser_map({ delta: true })` after the click to see ONLY what changed in the DOM.
+3. **Wait**: If elements aren't ready, use `browser_wait` for network idle rather than arbitrary sleep timeouts.

package/general/table-scraping.md

@@ -0,0 +1,12 @@
+---
+id: table-scraping
+tier: general
+triggers: []
+tools: []
+weavetab: ">=2.5.0"
+---
+# Table Scraping Pattern
+
+1. **Locate**: Use `browser_map({ prune: true })` to locate the table container.
+2. **Scrape**: Do not manually click through pages or read the DOM raw. Use `browser_scrape` targeting the table structure to extract row data cleanly into JSON.
+3. **Pagination**: If scraping multiple pages, map the "Next" button ID once, then loop: `browser_scrape` -> `browser_click` (Next) -> verify change -> repeat.

package/package.json

@@ -0,0 +1,42 @@
+{
+  "name": "@weavetab/skills",
+  "version": "1.0.0",
+  "description": "Supercharge your AI Agents with Weavetab MCP Skills. Native drop-in skills for Antigravity, Cursor, Cline, and more.",
+  "author": "fy2ne",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/Weavetab/skills.git"
+  },
+  "homepage": "https://github.com/Weavetab/skills#readme",
+  "bugs": {
+    "url": "https://github.com/Weavetab/skills/issues"
+  },
+  "keywords": [
+    "weavetab",
+    "weavetab-mcp",
+    "mcp",
+    "ai-agent",
+    "cursor",
+    "cline",
+    "antigravity",
+    "browser-automation",
+    "agent-skills",
+    "openclaw"
+  ],
+  "peerDependenciesMeta": {
+    "weavetab": {
+      "optional": true
+    }
+  },
+  "main": "./src/api.js",
+  "bin": {
+    "weavetab-skills": "./bin/cli.js"
+  },
+  "scripts": {
+    "build": "node scripts/build.js"
+  },
+  "dependencies": {
+    "@inquirer/prompts": "^7.3.1",
+    "yaml": "^2.3.4"
+  }
+}

package/scripts/build.js

@@ -0,0 +1,48 @@
+const fs = require('fs');
+const path = require('path');
+const yaml = require('yaml');
+
+const dirs = ['general', 'automation', 'advanced'];
+const skills = [];
+
+function parseFile(filePath) {
+  const content = fs.readFileSync(filePath, 'utf8');
+  const match = content.match(/^---\r?\n([\s\S]*?)\r?\n---/);
+  if (!match) return null;
+  
+  const frontmatter = yaml.parse(match[1]);
+  
+  // Convert absolute path to relative path
+  const relativePath = path.relative(path.join(__dirname, '..'), filePath);
+  
+  return {
+    ...frontmatter,
+    filePath: relativePath.replace(/\\/g, '/')
+  };
+}
+
+for (const dir of dirs) {
+  const dirPath = path.join(__dirname, '..', dir);
+  if (!fs.existsSync(dirPath)) continue;
+  
+  const files = fs.readdirSync(dirPath).filter(f => f.endsWith('.md') && f !== 'README.md');
+  for (const file of files) {
+    const fullPath = path.join(dirPath, file);
+    try {
+      const parsed = parseFile(fullPath);
+      if (parsed) {
+        skills.push(parsed);
+      }
+    } catch (e) {
+      console.error(`Error parsing ${fullPath}: ${e.message}`);
+      process.exit(1);
+    }
+  }
+}
+
+fs.writeFileSync(
+  path.join(__dirname, '..', 'skills.json'), 
+  JSON.stringify({ skills }, null, 2)
+);
+
+console.log(`Successfully built skills.json with ${skills.length} skills.`);

package/skills.json

@@ -0,0 +1,84 @@
+{
+  "skills": [
+    {
+      "id": "captcha-evasion",
+      "tier": "general",
+      "triggers": [],
+      "tools": [],
+      "weavetab": ">=2.5.0",
+      "filePath": "general/captcha-evasion.md"
+    },
+    {
+      "id": "file-upload",
+      "tier": "general",
+      "triggers": [],
+      "tools": [],
+      "weavetab": ">=2.5.0",
+      "filePath": "general/file-upload.md"
+    },
+    {
+      "id": "form-filling",
+      "tier": "general",
+      "triggers": [],
+      "tools": [],
+      "weavetab": ">=2.5.0",
+      "filePath": "general/form-filling.md"
+    },
+    {
+      "id": "guardrails",
+      "tier": "general",
+      "triggers": [],
+      "tools": [],
+      "weavetab": ">=2.5.0",
+      "filePath": "general/guardrails.md"
+    },
+    {
+      "id": "iframe-handling",
+      "tier": "general",
+      "triggers": [],
+      "tools": [],
+      "weavetab": ">=2.5.0",
+      "filePath": "general/iframe-handling.md"
+    },
+    {
+      "id": "infinite-scroll",
+      "tier": "general",
+      "triggers": [],
+      "tools": [],
+      "weavetab": ">=2.5.0",
+      "filePath": "general/infinite-scroll.md"
+    },
+    {
+      "id": "login-flow",
+      "tier": "general",
+      "triggers": [],
+      "tools": [],
+      "weavetab": ">=2.5.0",
+      "filePath": "general/login-flow.md"
+    },
+    {
+      "id": "multi-tab-management",
+      "tier": "general",
+      "triggers": [],
+      "tools": [],
+      "weavetab": ">=2.5.0",
+      "filePath": "general/multi-tab-management.md"
+    },
+    {
+      "id": "spa-navigation",
+      "tier": "general",
+      "triggers": [],
+      "tools": [],
+      "weavetab": ">=2.5.0",
+      "filePath": "general/spa-navigation.md"
+    },
+    {
+      "id": "table-scraping",
+      "tier": "general",
+      "triggers": [],
+      "tools": [],
+      "weavetab": ">=2.5.0",
+      "filePath": "general/table-scraping.md"
+    }
+  ]
+}

package/src/api.js

@@ -0,0 +1,128 @@
+const fs = require('fs');
+const path = require('path');
+
+/**
+ * Programmatically install Weavetab MCP skills into a target project.
+ * 
+ * @param {Object} options
+ * @param {string} options.targetFramework - '.agents', '.cursor', '.clinerules', or 'generic'
+ * @param {string[]} options.categories - Array of categories to install (e.g. ['general'])
+ * @param {string} options.projectRoot - The root directory to install into (defaults to process.cwd())
+ * @returns {Promise<string>} - A message indicating success and the path where skills were installed.
+ */
+async function installSkills({ targetFramework, categories, projectRoot = process.cwd() }) {
+  const packageRoot = path.join(__dirname, '..');
+  
+  if (!categories || categories.length === 0) {
+    throw new Error("No categories selected for installation.");
+  }
+
+  // Gather all markdown files from the selected categories
+  let allMarkdownFiles = [];
+  for (const category of categories) {
+    const categoryDir = path.join(packageRoot, category);
+    if (!fs.existsSync(categoryDir)) continue;
+    
+    const files = fs.readdirSync(categoryDir).filter(f => f.endsWith('.md'));
+    for (const file of files) {
+      allMarkdownFiles.push({
+        name: file,
+        content: fs.readFileSync(path.join(categoryDir, file), 'utf8')
+      });
+    }
+  }
+
+  if (allMarkdownFiles.length === 0) {
+    throw new Error("No skills found in the selected categories.");
+  }
+
+  // Process and inject based on the framework
+  if (targetFramework === '.clinerules') {
+    // Cline uses a single file at the root
+    const clineFile = path.join(projectRoot, '.clinerules');
+    let combinedContent = "\n\n# --- Weavetab MCP Skills ---\n\n";
+    for (const file of allMarkdownFiles) {
+      combinedContent += `## ${file.name}\n${file.content}\n\n`;
+    }
+    fs.appendFileSync(clineFile, combinedContent);
+    return `Successfully appended ${allMarkdownFiles.length} skills to ${clineFile}`;
+  } 
+  
+  else if (targetFramework === '.cursor') {
+    // Cursor uses .mdc files in .cursor/rules
+    const cursorRulesDir = path.join(projectRoot, '.cursor', 'rules');
+    fs.mkdirSync(cursorRulesDir, { recursive: true });
+    
+    for (const file of allMarkdownFiles) {
+      const destFile = path.join(cursorRulesDir, file.name.replace('.md', '.mdc'));
+      fs.writeFileSync(destFile, file.content);
+    }
+    return `Successfully installed ${allMarkdownFiles.length} skills to ${cursorRulesDir}`;
+  } 
+  
+  else if (targetFramework === '.agents') {
+    // Standard agent customizations folder
+    const agentsDir = path.join(projectRoot, '.agents', 'skills', 'weavetab');
+    fs.mkdirSync(agentsDir, { recursive: true });
+    
+    for (const file of allMarkdownFiles) {
+      fs.writeFileSync(path.join(agentsDir, file.name), file.content);
+    }
+    return `Successfully installed ${allMarkdownFiles.length} skills to ${agentsDir}`;
+  } 
+  
+  else if (targetFramework === '.openclaw') {
+    // OpenClaw AI folder
+    const openClawDir = path.join(projectRoot, '.openclaw', 'skills');
+    fs.mkdirSync(openClawDir, { recursive: true });
+    
+    for (const file of allMarkdownFiles) {
+      fs.writeFileSync(path.join(openClawDir, file.name), file.content);
+    }
+    return `Successfully installed ${allMarkdownFiles.length} skills to ${openClawDir}`;
+  } 
+  
+  else if (targetFramework === '.roocode') {
+    const rooDir = path.join(projectRoot, '.roocode', 'rules');
+    fs.mkdirSync(rooDir, { recursive: true });
+    
+    for (const file of allMarkdownFiles) {
+      fs.writeFileSync(path.join(rooDir, file.name), file.content);
+    }
+    return `Successfully installed ${allMarkdownFiles.length} skills to ${rooDir}`;
+  } 
+  
+  else if (targetFramework === '.windsurf') {
+    const windsurfDir = path.join(projectRoot, '.windsurf', 'rules');
+    fs.mkdirSync(windsurfDir, { recursive: true });
+    
+    for (const file of allMarkdownFiles) {
+      fs.writeFileSync(path.join(windsurfDir, file.name), file.content);
+    }
+    return `Successfully installed ${allMarkdownFiles.length} skills to ${windsurfDir}`;
+  } 
+  
+  else if (targetFramework === '.aider') {
+    const aiderFile = path.join(projectRoot, '.aider.conf.yml');
+    let combinedContent = "\n# --- Weavetab MCP Skills ---\n";
+    for (const file of allMarkdownFiles) {
+      // Very basic formatting for appending to Aider conf
+      combinedContent += `# ${file.name}\n# ${file.content.split('\n').join('\n# ')}\n\n`;
+    }
+    fs.appendFileSync(aiderFile, combinedContent);
+    return `Successfully appended ${allMarkdownFiles.length} skills to ${aiderFile}`;
+  } 
+  
+  else {
+    // Generic folder
+    const genericDir = path.join(projectRoot, 'weavetab-skills');
+    fs.mkdirSync(genericDir, { recursive: true });
+    
+    for (const file of allMarkdownFiles) {
+      fs.writeFileSync(path.join(genericDir, file.name), file.content);
+    }
+    return `Successfully installed ${allMarkdownFiles.length} skills to ${genericDir}`;
+  }
+}
+
+module.exports = { installSkills };