[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M72T2WW8KAQASN8A72B6EM84":3,"skill-same:01M72T2WW8KAQASN8A72B6EM84":52},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":31,"commitSha":32,"skillDirectory":33,"forksCount":14,"repositoryHtmlUrl":34,"repositoryHomepageUrl":34,"ownerHtmlUrl":35,"ownerBlogUrl":6,"ownerDescription":36,"updatedAt":37,"install":38,"resources":51},"01M72T2WW8KAQASN8A72B6EM84","gemini-deep-research",null,"Advanced deep research agent using Gemini 3.1 Pro with customtools endpoint for optimal agent performance. Deep research, thinking mode, and project management for 50+ SWAL projects. Updated 2026-04-20.","iberi22\u002Fskills","iberi22","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F10615454?v=4","01M72SQ2NXAA3GW0FSH053A5ES","Brahyan Belalcazar","01M72SQ2PYCH0SSA1K61NGEHDV",0,"MIT","https:\u002F\u002Fgithub.com\u002Fiberi22\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fgeneral\u002Fgemini-deep-research\u002FSKILL.md",[],[],"2026-11-10 22:39:42",false,"---\nname: gemini-deep-research\ndescription: Advanced deep research agent using Gemini 3.1 Pro with customtools endpoint for optimal agent performance. Deep research, thinking mode, and project management for 50+ SWAL projects. Updated 2026-04-20.\nversion: \"1.0.0\"\nupdated: \"2026-04-23\"\nauthor: swal\nlicense: MIT\nmetadata:\n  provider: Google AI Studio\n  model: gemini-3.1-pro-preview\n  features: thinking, code-execution, google-search, customtools\n---\n\n# SWAL Deep Research Agent - Gemini 3.1 Pro Edition\n\n> **🧠 Most Advanced Research System**\n> Using Gemini 3.1 Pro with customtools endpoint\n\n---\n\n## 🤖 Model Selection (UPDATED)\n\n### ⭐ PRIMARY: `gemini-3.1-pro-preview-customtools`\n- **Released**: February 19, 2026\n- **Why**: Optimized for agent tool use - prioritizes custom tools over bash\n- **Endpoint**: Native Google API (not OpenAI compatible)\n- **Cost**: Same as standard\n\n### BACKUP: `gemini-2.5-pro`\n- **Why**: Proven, stable\n- **Endpoint**: OpenAI compatible available\n\n---\n\n## 🔑 API Keys (4 Keys Available)\n\n```\nKeys: AIzaSyBXo..., AIzaSyCxp..., AIzaSyDtx..., AIzaSyDj9...\n```\n\n---\n\n## 🚨 IMPORTANT: API Endpoint\n\n**Gemini 3.1 uses NATIVE Google API, NOT OpenAI compatible:**\n\n```javascript\n\u002F\u002F Native Google API (REQUIRED for 3.1)\nPOST https:\u002F\u002Fgenerativelanguage.googleapis.com\u002Fv1beta\u002Fmodels\u002Fgemini-3.1-pro-preview:generateContent\n\n\u002F\u002F NOT the OpenAI endpoint!\nPOST https:\u002F\u002Fgenerativelanguage.googleapis.com\u002Fv1beta\u002Fopenai\u002Fchat\u002Fcompletions  \u002F\u002F ❌ Won't work\n```\n\n---\n\n## 📡 API Configuration\n\n### Native Google API (for 3.1 Pro)\n```javascript\nconst response = await fetch(\n  `https:\u002F\u002Fgenerativelanguage.googleapis.com\u002Fv1beta\u002Fmodels\u002Fgemini-3.1-pro-preview:generateContent?key=${API_KEY}`,\n  {\n    method: 'POST',\n    headers: { 'Content-Type': 'application\u002Fjson' },\n    body: JSON.stringify({\n      contents: [{ parts: [{ text: prompt }] }],\n      generationConfig: {\n        maxOutputTokens: 8192,\n        temperature: 0.2\n      }\n    })\n  }\n);\n```\n\n### OpenAI Compatible (for 2.5 Flash)\n```javascript\n\u002F\u002F Works with OpenAI-compatible apps\nPOST https:\u002F\u002Fgenerativelanguage.googleapis.com\u002Fv1beta\u002Fopenai\u002Fchat\u002Fcompletions\n```\n\n---\n\n## 🛠 CustomTools Endpoint\n\n### What it does:\n- **Better tool prioritization**: Uses custom tools (view_file, search_code) instead of generic bash\n- **Less hallucinations**: More reliable file operations\n- **Agent-optimized**: Built for coding agents\n\n### Function Declarations (Recommended Names)\n```javascript\n{\n  \"tools\": [{\n    \"function_declarations\": [\n      {\n        \"name\": \"view_file\",\n        \"description\": \"Read a file from the filesystem\",\n        \"parameters\": {\n          \"type\": \"object\",\n          \"properties\": {\n            \"path\": { \"type\": \"string\" }\n          }\n        }\n      },\n      {\n        \"name\": \"search_code\",\n        \"description\": \"Search for code patterns\",\n        \"parameters\": {\n          \"type\": \"object\",\n          \"properties\": {\n            \"query\": { \"type\": \"string\" }\n          }\n        }\n      },\n      {\n        \"name\": \"execute_bash\",\n        \"description\": \"Execute a bash command\",\n        \"parameters\": {\n          \"type\": \"object\",\n          \"properties\": {\n            \"command\": { \"type\": \"string\" }\n          }\n        }\n      }\n    ]\n  }]\n}\n```\n\n---\n\n## 🔍 Available Models on Your Keys\n\n| Model | Type | Available |\n|-------|------|-----------|\n| `gemini-3.1-pro-preview` | Reasoning | ✅ |\n| `gemini-3.1-pro-preview-customtools` | Agent Tools | ✅ |\n| `gemini-3.1-flash-lite-preview` | Fast | ✅ |\n| `gemini-2.5-pro` | Reasoning | ✅ |\n| `gemini-2.5-flash` | Balance | ✅ |\n\n---\n\n## 🧠 Thinking Mode\n\nGemini 3.1 has enhanced thinking:\n\n```javascript\n{\n  \"model\": \"gemini-3.1-pro-preview\",\n  \"contents\": {\n    \"parts\": [{ \"text\": \"Your complex research query\" }]\n  },\n  \"generationConfig\": {\n    \"thinkingConfig\": {\n      \"thinkingBudgetTokens\": 16384\n    }\n  }\n}\n```\n\n---\n\n## 🔄 Auto-Research Loop\n\n```\n┌─────────────────────────────────────────────────────────┐\n│                   DEEP RESEARCH LOOP                     │\n├─────────────────────────────────────────────────────────┤\n│                                                          │\n│  1. Gemini 3.1 analyzes query                           │\n│  2. Generates search strategy                           │\n│  3. Uses google_search for web grounding               │\n│  4. Synthesizes findings                               │\n│  5. Identifies gaps                                     │\n│  6. Iterates if needed                                  │\n│  7. Final comprehensive report                          │\n│                                                          │\n└─────────────────────────────────────────────────────────┘\n```\n\n### Research Prompt Template\n```javascript\nconst researchPrompt = `\nPERFORM DEEP RESEARCH on: {TOPIC}\n\nRequirements:\n1. Use google_search for current information\n2. Cite sources with URLs\n3. Identify knowledge gaps\n4. Provide actionable recommendations\n\nOutput: Comprehensive Markdown report\n`;\n```\n\n---\n\n## 📊 Rate Limit Strategy (4 Keys)\n\n```javascript\nclass ResearchRouter {\n  constructor() {\n    this.keys = [\n      '${GEMINI_API_KEY}',\n      '${GEMINI_API_KEY}',\n      '${GEMINI_API_KEY}',\n      '${GEMINI_API_KEY}'\n    ];\n    this.index = 0;\n  }\n\n  getKey() {\n    const key = this.keys[this.index];\n    this.index = (this.index + 1) % this.keys.length;\n    return key;\n  }\n}\n```\n\n---\n\n## 🎯 Model Selection Guide\n\n| Task | Model | Why |\n|------|-------|-----|\n| **Coding Agent** | 3.1-pro-customtools | Best tool use |\n| **Deep Research** | 3.1-pro-preview | Best reasoning |\n| **Quick Analysis** | 2.5-flash | Fast, cheap |\n| **Long Context** | 3.1-pro (1M) | 1M tokens |\n\n---\n\n## 🚀 Quick Usage\n\n### Deep Research (Native API)\n```bash\n# Use the research script with 3.1\nnode scripts\u002Fdeep-research-loop.js \"AI agent frameworks 2026\"\n```\n\n### Coding Agent (CustomTools)\n```bash\n# Use OpenCode with customtools model\nopencode run --model google\u002Fgemini-3.1-pro-preview-customtools \"Create a React component\"\n```\n\n### Quick Analysis (OpenAI Compatible)\n```bash\n# Use Gemini CLI\ngemini -p \"Analyze this code architecture\" -m gemini-2.5-flash\n```\n\n---\n\n## 📁 Files\n\n- `scripts\u002Fdeep-research-loop.js` - Auto-research\n- `scripts\u002Fresearch-router.js` - Key rotation\n- `scripts\u002Fswal-project-manager.js` - Project tracking\n\n---\n\n**Version:** 2.0 (Gemini 3.1 Pro)\n**Updated:** 2026-04-20\n**Status:** ✅ Ready with Gemini 3.1 Pro","# SWAL Deep Research Agent - Gemini 3.1 Pro Edition\n\n> **🧠 Most Advanced Research System**\n> Using Gemini 3.1 Pro with customtools endpoint\n\n---\n\n## 🤖 Model Selection (UPDATED)\n\n### ⭐ PRIMARY: `gemini-3.1-pro-preview-customtools`\n- **Released**: February 19, 2026\n- **Why**: Optimized for agent tool use - prioritizes custom tools over bash\n- **Endpoint**: Native Google API (not OpenAI compatible)\n- **Cost**: Same as standard\n\n### BACKUP: `gemini-2.5-pro`\n- **Why**: Proven, stable\n- **Endpoint**: OpenAI compatible available\n\n---\n\n## 🔑 API Keys (4 Keys Available)\n\n```\nKeys: AIzaSyBXo..., AIzaSyCxp..., AIzaSyDtx..., AIzaSyDj9...\n```\n\n---\n\n## 🚨 IMPORTANT: API Endpoint\n\n**Gemini 3.1 uses NATIVE Google API, NOT OpenAI compatible:**\n\n```javascript\n\u002F\u002F Native Google API (REQUIRED for 3.1)\nPOST https:\u002F\u002Fgenerativelanguage.googleapis.com\u002Fv1beta\u002Fmodels\u002Fgemini-3.1-pro-preview:generateContent\n\n\u002F\u002F NOT the OpenAI endpoint!\nPOST https:\u002F\u002Fgenerativelanguage.googleapis.com\u002Fv1beta\u002Fopenai\u002Fchat\u002Fcompletions  \u002F\u002F ❌ Won't work\n```\n\n---\n\n## 📡 API Configuration\n\n### Native Google API (for 3.1 Pro)\n```javascript\nconst response = await fetch(\n  `https:\u002F\u002Fgenerativelanguage.googleapis.com\u002Fv1beta\u002Fmodels\u002Fgemini-3.1-pro-preview:generateContent?key=${API_KEY}`,\n  {\n    method: 'POST',\n    headers: { 'Content-Type': 'application\u002Fjson' },\n    body: JSON.stringify({\n      contents: [{ parts: [{ text: prompt }] }],\n      generationConfig: {\n        maxOutputTokens: 8192,\n        temperature: 0.2\n      }\n    })\n  }\n);\n```\n\n### OpenAI Compatible (for 2.5 Flash)\n```javascript\n\u002F\u002F Works with OpenAI-compatible apps\nPOST https:\u002F\u002Fgenerativelanguage.googleapis.com\u002Fv1beta\u002Fopenai\u002Fchat\u002Fcompletions\n```\n\n---\n\n## 🛠 CustomTools Endpoint\n\n### What it does:\n- **Better tool prioritization**: Uses custom tools (view_file, search_code) instead of generic bash\n- **Less hallucinations**: More reliable file operations\n- **Agent-optimized**: Built for coding agents\n\n### Function Declarations (Recommended Names)\n```javascript\n{\n  \"tools\": [{\n    \"function_declarations\": [\n      {\n        \"name\": \"view_file\",\n        \"description\": \"Read a file from the filesystem\",\n        \"parameters\": {\n          \"type\": \"object\",\n          \"properties\": {\n            \"path\": { \"type\": \"string\" }\n          }\n        }\n      },\n      {\n        \"name\": \"search_code\",\n        \"description\": \"Search for code patterns\",\n        \"parameters\": {\n          \"type\": \"object\",\n          \"properties\": {\n            \"query\": { \"type\": \"string\" }\n          }\n        }\n      },\n      {\n        \"name\": \"execute_bash\",\n        \"description\": \"Execute a bash command\",\n        \"parameters\": {\n          \"type\": \"object\",\n          \"properties\": {\n            \"command\": { \"type\": \"string\" }\n          }\n        }\n      }\n    ]\n  }]\n}\n```\n\n---\n\n## 🔍 Available Models on Your Keys\n\n| Model | Type | Available |\n|-------|------|-----------|\n| `gemini-3.1-pro-preview` | Reasoning | ✅ |\n| `gemini-3.1-pro-preview-customtools` | Agent Tools | ✅ |\n| `gemini-3.1-flash-lite-preview` | Fast | ✅ |\n| `gemini-2.5-pro` | Reasoning | ✅ |\n| `gemini-2.5-flash` | Balance | ✅ |\n\n---\n\n## 🧠 Thinking Mode\n\nGemini 3.1 has enhanced thinking:\n\n```javascript\n{\n  \"model\": \"gemini-3.1-pro-preview\",\n  \"contents\": {\n    \"parts\": [{ \"text\": \"Your complex research query\" }]\n  },\n  \"generationConfig\": {\n    \"thinkingConfig\": {\n      \"thinkingBudgetTokens\": 16384\n    }\n  }\n}\n```\n\n---\n\n## 🔄 Auto-Research Loop\n\n```\n┌─────────────────────────────────────────────────────────┐\n│                   DEEP RESEARCH LOOP                     │\n├─────────────────────────────────────────────────────────┤\n│                                                          │\n│  1. Gemini 3.1 analyzes query                           │\n│  2. Generates search strategy                           │\n│  3. Uses google_search for web grounding               │\n│  4. Synthesizes findings                               │\n│  5. Identifies gaps                                     │\n│  6. Iterates if needed                                  │\n│  7. Final comprehensive report                          │\n│                                                          │\n└─────────────────────────────────────────────────────────┘\n```\n\n### Research Prompt Template\n```javascript\nconst researchPrompt = `\nPERFORM DEEP RESEARCH on: {TOPIC}\n\nRequirements:\n1. Use google_search for current information\n2. Cite sources with URLs\n3. Identify knowledge gaps\n4. Provide actionable recommendations\n\nOutput: Comprehensive Markdown report\n`;\n```\n\n---\n\n## 📊 Rate Limit Strategy (4 Keys)\n\n```javascript\nclass ResearchRouter {\n  constructor() {\n    this.keys = [\n      '${GEMINI_API_KEY}',\n      '${GEMINI_API_KEY}',\n      '${GEMINI_API_KEY}',\n      '${GEMINI_API_KEY}'\n    ];\n    this.index = 0;\n  }\n\n  getKey() {\n    const key = this.keys[this.index];\n    this.index = (this.index + 1) % this.keys.length;\n    return key;\n  }\n}\n```\n\n---\n\n## 🎯 Model Selection Guide\n\n| Task | Model | Why |\n|------|-------|-----|\n| **Coding Agent** | 3.1-pro-customtools | Best tool use |\n| **Deep Research** | 3.1-pro-preview | Best reasoning |\n| **Quick Analysis** | 2.5-flash | Fast, cheap |\n| **Long Context** | 3.1-pro (1M) | 1M tokens |\n\n---\n\n## 🚀 Quick Usage\n\n### Deep Research (Native API)\n```bash\n# Use the research script with 3.1\nnode scripts\u002Fdeep-research-loop.js \"AI agent frameworks 2026\"\n```\n\n### Coding Agent (CustomTools)\n```bash\n# Use OpenCode with customtools model\nopencode run --model google\u002Fgemini-3.1-pro-preview-customtools \"Create a React component\"\n```\n\n### Quick Analysis (OpenAI Compatible)\n```bash\n# Use Gemini CLI\ngemini -p \"Analyze this code architecture\" -m gemini-2.5-flash\n```\n\n---\n\n## 📁 Files\n\n- `scripts\u002Fdeep-research-loop.js` - Auto-research\n- `scripts\u002Fresearch-router.js` - Key rotation\n- `scripts\u002Fswal-project-manager.js` - Project tracking\n\n---\n\n**Version:** 2.0 (Gemini 3.1 Pro)\n**Updated:** 2026-04-20\n**Status:** ✅ Ready with Gemini 3.1 Pro",{"name":5,"author":24,"license":15,"updated":25,"version":26,"metadata":27,"description":7},"swal","2026-04-23","1.0.0",{"model":28,"features":29,"provider":30},"gemini-3.1-pro-preview","thinking, code-execution, google-search, customtools","Google AI Studio","main","688810f2886d212632acf81e532b24e483dd8b98","skills\u002Fgeneral\u002Fgemini-deep-research","https:\u002F\u002Fgithub.com\u002Fiberi22\u002Fskills","https:\u002F\u002Fgithub.com\u002Fiberi22","Senior WEB Full-Stack","2026-10-12 02:46:21",{"source":8,"skillName":5,"commands":39,"prompt":49,"zipAvailable":50},[40,43,46],{"tool":41,"command":42},"npx","npx skills add iberi22\u002Fskills --skill gemini-deep-research",{"tool":44,"command":45},"bunx","bunx skills add iberi22\u002Fskills --skill gemini-deep-research",{"tool":47,"command":48},"pnpm","pnpm dlx skills add iberi22\u002Fskills --skill gemini-deep-research","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fiberi22\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fgeneral\u002Fgemini-deep-research\u002FSKILL.md",true,[],[53,59,66,72,78,84],{"skillNo":54,"name":55,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":19},"01M72TWQEQD46ZC9TRBAABF78P","web-design-guidelines","https:\u002F\u002Fgithub.com\u002Fiberi22\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fweb\u002Fweb-design-guidelines\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M72TWMQFTYD08DGXVAXRPPEB","blocked-page-recovery","Use when a fetch fails: 403\u002F429, paywall, WAF, bot wall. Recovers the page via archives and provenance-checked fallbacks.","https:\u002F\u002Fgithub.com\u002Fiberi22\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fweb\u002Fblocked-page-recovery\u002FSKILL.md",[],[],{"skillNo":67,"name":68,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M72TVQMS3HCTX18C8T97WTQQ","studio-coach","https:\u002F\u002Fgithub.com\u002Fiberi22\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Ftools\u002Fstudio-coach\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M72TVNA2PR9RZW0ZAXYP1G6J","src-generator","https:\u002F\u002Fgithub.com\u002Fiberi22\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Ftools\u002Fsrc-generator\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M72TVEJHHB6SAT3R3TY9MJKS","sqlite-pro","https:\u002F\u002Fgithub.com\u002Fiberi22\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Ftools\u002Fsqlite-pro\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M72TVC2EVBHQHCX96VTTYKB8","rigor-evaluator","Use when you need a quick heuristic rigor score (10 criteria, 0-2 each) for the features in a .gitcore\u002Ffeatures.json file, to decide which features are ready to ship.","https:\u002F\u002Fgithub.com\u002Fiberi22\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Ftools\u002Frigor-evaluator\u002FSKILL.md",[],[]]