[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AYM93D09RYCPP4XZG1AR7Q":3,"skill-same:01M7AYM93D09RYCPP4XZG1AR7Q":67},{"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":44,"commitSha":45,"skillDirectory":46,"forksCount":47,"repositoryHtmlUrl":48,"repositoryHomepageUrl":6,"ownerHtmlUrl":49,"ownerBlogUrl":50,"ownerDescription":51,"updatedAt":52,"install":53,"resources":66},"01M7AYM93D09RYCPP4XZG1AR7Q","web-research",null,"Fetch docs via curl when browser tools unavailable.","oat431\u002Fhermes_config_backup","oat431","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F54722120?v=4","01M7AYD07EH8SYPMYY8CGZKS22","Oralita","01M7AYD0816QDF77E3TWCRSQC9",1,"MIT","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Ffull-stack\u002Fweb-research\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: web-research\ndescription: \"Fetch docs via curl when browser tools unavailable.\"\nversion: 1.0.0\nauthor: Hermes Agent\nlicense: MIT\nplatforms: [linux, macos, windows]\nmetadata:\n  hermes:\n    tags: [Research, Documentation, Web, Technical, APIs, Frameworks]\n    category: research\n    related_skills: [grounded-citations, arxiv, blogwatcher]\n---\n\n# Web Research\n\nFetch and synthesize technical documentation from official sources when browser automation tools fail or are unavailable. Covers API documentation, framework guides, release notes, and technical references from sites like Microsoft Learn, GitHub, Stack Overflow, and vendor documentation.\n\n## When to Use\n\n- Researching APIs, frameworks, or libraries (e.g., \".NET 9 features\", \"React hooks\", \"gRPC in Go\")\n- Fetching release notes, migration guides, or breaking changes\n- Gathering technical specifications or deployment patterns\n- Synthesizing information from multiple official documentation sources\n- When `browser_navigate` fails due to missing Chrome\u002FPuppeteer\u002FPlaywright\n\nSkip for: casual web searches (use `web_search`), academic papers (use `arxiv`), blog monitoring (use `blogwatcher`), or when you need interactive browser automation (use `computer_use`).\n\n## Prerequisites\n\n- `curl` (available on all platforms)\n- Python 3 with `re` module (stdlib)\n- `terminal` tool for shell commands\n\n## Browser Fallback Pattern\n\nWhen `browser_navigate` fails with \"Chrome not found\" or similar errors, fall back to curl + regex-based HTML text extraction:\n\n```python\nfrom hermes_tools import terminal\nimport re\n\ndef extract_text(html):\n    \"\"\"Strip HTML tags, remove script\u002Fstyle blocks, normalize whitespace.\"\"\"\n    html = re.sub(r'\u003Cscript[^>]*>.*?\u003C\u002Fscript>', '', html, flags=re.DOTALL | re.IGNORECASE)\n    html = re.sub(r'\u003Cstyle[^>]*>.*?\u003C\u002Fstyle>', '', html, flags=re.DOTALL | re.IGNORECASE)\n    html = re.sub(r'\u003C[^>]+>', ' ', html)\n    html = re.sub(r'\\s+', ' ', html)\n    return html.strip()\n\n# Fetch documentation\nr = terminal(\"curl -sL '\u003Curl>' -H 'User-Agent: Mozilla\u002F5.0' 2>\u002Fdev\u002Fnull\")\ntext = extract_text(r['output'])\nprint(text[:5000])  # Adjust slice based on content length\n```\n\n**Why this works:**\n- Many documentation sites (Microsoft Learn, GitHub, etc.) are server-rendered, not SPA\u002FJavaScript-heavy\n- curl with a User-Agent header bypasses most bot detection\n- Regex extraction is faster than full HTML parsing for simple text extraction\n- Works in headless environments where Chrome isn't installed\n\n**Limitations:**\n- Won't work for JavaScript-rendered content (React\u002FVue\u002FAngular apps)\n- May miss content loaded via AJAX\u002Ffetch calls\n- Complex table layouts or nested structures may lose formatting\n\n## Procedure\n\n① **Identify official sources first.** For framework\u002Flibrary research, start with:\n- Official documentation (learn.microsoft.com, docs.python.org, etc.)\n- GitHub repositories (README, docs\u002F, release notes)\n- RFC\u002Fspecification documents\n- Vendor blogs or engineering posts\n\n② **Batch fetch in parallel.** Use `execute_code` to fetch multiple URLs concurrently:\n\n```python\nfrom hermes_tools import terminal\nimport re\n\ndef extract_text(html):\n    # ... (function from above)\n\nurls = [\n    ('https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Fdotnet\u002Fcore\u002Fwhats-new\u002Fdotnet-9\u002Foverview', 'dotnet9'),\n    ('https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Faspnet\u002Fcore\u002Frelease-notes\u002Faspnetcore-9.0', 'aspnet9'),\n]\n\nfor url, key in urls:\n    r = terminal(f\"curl -sL '{url}' -H 'User-Agent: Mozilla\u002F5.0' 2>\u002Fdev\u002Fnull\")\n    text = extract_text(r['output'])\n    print(f\"=== {key.upper()} ===\")\n    print(text[:4000])\n    print(\"\\n---\\n\")\n```\n\n③ **Extract and synthesize.** Read the fetched text, identify key sections (features, breaking changes, code examples), and synthesize into a structured summary. Focus on:\n- New features or capabilities\n- Breaking changes or deprecations\n- Performance improvements\n- Code examples or patterns\n- Migration paths\n\n④ **Cite sources.** If using the `grounded-citations` skill, register URLs and cite inline. Otherwise, append a `Sources:` section with URLs.\n\n⑤ **Structure the output.** For comprehensive research, organize by topic:\n\n```markdown\n## Topic 1: Feature Name\n- Key points\n- Code examples\n- Breaking changes\n\n## Topic 2: Deployment Patterns\n- Docker multi-stage builds\n- CI\u002FCD workflows\n```\n\n## Pitfalls\n\n- **JavaScript-rendered sites.** If curl returns empty or minimal content, the site likely requires JavaScript execution. Fall back to `browser_navigate` (if available) or find alternative documentation sources (GitHub READMEs, cached versions, etc.).\n- **Rate limiting.** Some sites throttle curl requests. Add delays between fetches or use different User-Agent strings if you hit 429 errors.\n- **HTML entity encoding.** Extracted text may contain `&amp;`, `&lt;`, `&gt;`. Use `html.unescape()` if you need clean text:\n  ```python\n  import html\n  text = html.unescape(extract_text(r['output']))\n  ```\n- **Large pages.** Some documentation pages are 50k+ characters. Slice strategically (`text[:5000]`) or search for specific sections using regex.\n- **Outdated documentation.** Official docs may lag behind actual releases. Cross-reference with GitHub release notes or changelog files for the most current information.\n- **Missing context.** Extracted text loses visual structure (tables, code blocks, diagrams). Reconstruct formatting manually or note when structure is unclear.\n\n## Verification\n\nAfter synthesizing, verify by:\n1. Cross-checking key claims against 2+ sources when possible\n2. Checking GitHub repository for recent commits\u002Freleases\n3. Looking for \"last updated\" dates in documentation metadata\n4. Testing code examples if feasible (e.g., `dotnet new webapi --help` to verify CLI flags)\n\n## Example: .NET Feature Research\n\n```python\nfrom hermes_tools import terminal\nimport re\n\ndef extract_text(html):\n    html = re.sub(r'\u003Cscript[^>]*>.*?\u003C\u002Fscript>', '', html, flags=re.DOTALL | re.IGNORECASE)\n    html = re.sub(r'\u003Cstyle[^>]*>.*?\u003C\u002Fstyle>', '', html, flags=re.DOTALL | re.IGNORECASE)\n    html = re.sub(r'\u003C[^>]+>', ' ', html)\n    html = re.sub(r'\\s+', ' ', html)\n    return html.strip()\n\n# Fetch .NET 9 release notes\nr = terminal(\"curl -sL 'https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Fdotnet\u002Fcore\u002Fwhats-new\u002Fdotnet-9\u002Foverview' -H 'User-Agent: Mozilla\u002F5.0' 2>\u002Fdev\u002Fnull\")\ntext = extract_text(r['output'])\n\n# Search for specific sections\nimport re\nperformance_section = re.search(r'Performance.*?(?=##|$)', text, re.IGNORECASE | re.DOTALL)\nif performance_section:\n    print(performance_section.group(0)[:2000])\n```\n","# Web Research\n\nFetch and synthesize technical documentation from official sources when browser automation tools fail or are unavailable. Covers API documentation, framework guides, release notes, and technical references from sites like Microsoft Learn, GitHub, Stack Overflow, and vendor documentation.\n\n## When to Use\n\n- Researching APIs, frameworks, or libraries (e.g., \".NET 9 features\", \"React hooks\", \"gRPC in Go\")\n- Fetching release notes, migration guides, or breaking changes\n- Gathering technical specifications or deployment patterns\n- Synthesizing information from multiple official documentation sources\n- When `browser_navigate` fails due to missing Chrome\u002FPuppeteer\u002FPlaywright\n\nSkip for: casual web searches (use `web_search`), academic papers (use `arxiv`), blog monitoring (use `blogwatcher`), or when you need interactive browser automation (use `computer_use`).\n\n## Prerequisites\n\n- `curl` (available on all platforms)\n- Python 3 with `re` module (stdlib)\n- `terminal` tool for shell commands\n\n## Browser Fallback Pattern\n\nWhen `browser_navigate` fails with \"Chrome not found\" or similar errors, fall back to curl + regex-based HTML text extraction:\n\n```python\nfrom hermes_tools import terminal\nimport re\n\ndef extract_text(html):\n    \"\"\"Strip HTML tags, remove script\u002Fstyle blocks, normalize whitespace.\"\"\"\n    html = re.sub(r'\u003Cscript[^>]*>.*?\u003C\u002Fscript>', '', html, flags=re.DOTALL | re.IGNORECASE)\n    html = re.sub(r'\u003Cstyle[^>]*>.*?\u003C\u002Fstyle>', '', html, flags=re.DOTALL | re.IGNORECASE)\n    html = re.sub(r'\u003C[^>]+>', ' ', html)\n    html = re.sub(r'\\s+', ' ', html)\n    return html.strip()\n\n# Fetch documentation\nr = terminal(\"curl -sL '\u003Curl>' -H 'User-Agent: Mozilla\u002F5.0' 2>\u002Fdev\u002Fnull\")\ntext = extract_text(r['output'])\nprint(text[:5000])  # Adjust slice based on content length\n```\n\n**Why this works:**\n- Many documentation sites (Microsoft Learn, GitHub, etc.) are server-rendered, not SPA\u002FJavaScript-heavy\n- curl with a User-Agent header bypasses most bot detection\n- Regex extraction is faster than full HTML parsing for simple text extraction\n- Works in headless environments where Chrome isn't installed\n\n**Limitations:**\n- Won't work for JavaScript-rendered content (React\u002FVue\u002FAngular apps)\n- May miss content loaded via AJAX\u002Ffetch calls\n- Complex table layouts or nested structures may lose formatting\n\n## Procedure\n\n① **Identify official sources first.** For framework\u002Flibrary research, start with:\n- Official documentation (learn.microsoft.com, docs.python.org, etc.)\n- GitHub repositories (README, docs\u002F, release notes)\n- RFC\u002Fspecification documents\n- Vendor blogs or engineering posts\n\n② **Batch fetch in parallel.** Use `execute_code` to fetch multiple URLs concurrently:\n\n```python\nfrom hermes_tools import terminal\nimport re\n\ndef extract_text(html):\n    # ... (function from above)\n\nurls = [\n    ('https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Fdotnet\u002Fcore\u002Fwhats-new\u002Fdotnet-9\u002Foverview', 'dotnet9'),\n    ('https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Faspnet\u002Fcore\u002Frelease-notes\u002Faspnetcore-9.0', 'aspnet9'),\n]\n\nfor url, key in urls:\n    r = terminal(f\"curl -sL '{url}' -H 'User-Agent: Mozilla\u002F5.0' 2>\u002Fdev\u002Fnull\")\n    text = extract_text(r['output'])\n    print(f\"=== {key.upper()} ===\")\n    print(text[:4000])\n    print(\"\\n---\\n\")\n```\n\n③ **Extract and synthesize.** Read the fetched text, identify key sections (features, breaking changes, code examples), and synthesize into a structured summary. Focus on:\n- New features or capabilities\n- Breaking changes or deprecations\n- Performance improvements\n- Code examples or patterns\n- Migration paths\n\n④ **Cite sources.** If using the `grounded-citations` skill, register URLs and cite inline. Otherwise, append a `Sources:` section with URLs.\n\n⑤ **Structure the output.** For comprehensive research, organize by topic:\n\n```markdown\n## Topic 1: Feature Name\n- Key points\n- Code examples\n- Breaking changes\n\n## Topic 2: Deployment Patterns\n- Docker multi-stage builds\n- CI\u002FCD workflows\n```\n\n## Pitfalls\n\n- **JavaScript-rendered sites.** If curl returns empty or minimal content, the site likely requires JavaScript execution. Fall back to `browser_navigate` (if available) or find alternative documentation sources (GitHub READMEs, cached versions, etc.).\n- **Rate limiting.** Some sites throttle curl requests. Add delays between fetches or use different User-Agent strings if you hit 429 errors.\n- **HTML entity encoding.** Extracted text may contain `&amp;`, `&lt;`, `&gt;`. Use `html.unescape()` if you need clean text:\n  ```python\n  import html\n  text = html.unescape(extract_text(r['output']))\n  ```\n- **Large pages.** Some documentation pages are 50k+ characters. Slice strategically (`text[:5000]`) or search for specific sections using regex.\n- **Outdated documentation.** Official docs may lag behind actual releases. Cross-reference with GitHub release notes or changelog files for the most current information.\n- **Missing context.** Extracted text loses visual structure (tables, code blocks, diagrams). Reconstruct formatting manually or note when structure is unclear.\n\n## Verification\n\nAfter synthesizing, verify by:\n1. Cross-checking key claims against 2+ sources when possible\n2. Checking GitHub repository for recent commits\u002Freleases\n3. Looking for \"last updated\" dates in documentation metadata\n4. Testing code examples if feasible (e.g., `dotnet new webapi --help` to verify CLI flags)\n\n## Example: .NET Feature Research\n\n```python\nfrom hermes_tools import terminal\nimport re\n\ndef extract_text(html):\n    html = re.sub(r'\u003Cscript[^>]*>.*?\u003C\u002Fscript>', '', html, flags=re.DOTALL | re.IGNORECASE)\n    html = re.sub(r'\u003Cstyle[^>]*>.*?\u003C\u002Fstyle>', '', html, flags=re.DOTALL | re.IGNORECASE)\n    html = re.sub(r'\u003C[^>]+>', ' ', html)\n    html = re.sub(r'\\s+', ' ', html)\n    return html.strip()\n\n# Fetch .NET 9 release notes\nr = terminal(\"curl -sL 'https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Fdotnet\u002Fcore\u002Fwhats-new\u002Fdotnet-9\u002Foverview' -H 'User-Agent: Mozilla\u002F5.0' 2>\u002Fdev\u002Fnull\")\ntext = extract_text(r['output'])\n\n# Search for specific sections\nimport re\nperformance_section = re.search(r'Performance.*?(?=##|$)', text, re.IGNORECASE | re.DOTALL)\nif performance_section:\n    print(performance_section.group(0)[:2000])\n```\n",{"name":5,"author":24,"license":15,"version":25,"metadata":26,"platforms":40,"description":7},"Hermes Agent","1.0.0",{"hermes":27},{"tags":28,"category":35,"related_skills":36},[29,30,31,32,33,34],"Research","Documentation","Web","Technical","APIs","Frameworks","research",[37,38,39],"grounded-citations","arxiv","blogwatcher",[41,42,43],"linux","macos","windows","main","2988869c5a7d7be2225c9ea7a54a4a56a59b0d45","workflow\u002Ffull-stack\u002Fweb-research",0,"https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup","https:\u002F\u002Fgithub.com\u002Foat431","https:\u002F\u002Fpanomete.com\u002F","Hope and Luck is not a part of the strategy, \r\n\r\nSoftskill produces plan Hard skill produce product.","2026-11-03 00:43:08",{"source":8,"skillName":5,"commands":54,"prompt":64,"zipAvailable":65},[55,58,61],{"tool":56,"command":57},"npx","npx skills add oat431\u002Fhermes_config_backup --skill web-research",{"tool":59,"command":60},"bunx","bunx skills add oat431\u002Fhermes_config_backup --skill web-research",{"tool":62,"command":63},"pnpm","pnpm dlx skills add oat431\u002Fhermes_config_backup --skill web-research","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Ffull-stack\u002Fweb-research\u002FSKILL.md",true,[],[68,75,82,90,97,104],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M7AYZ2JY3NFKAPQSC39N5F3F","project-spec-docs","Fill\u002Fupdate project_spec docs (templates, MM handoffs).","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fui-ux\u002Fproject-spec-docs\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M7AYYRZPE7HQX7778558X52S","penpot-mcp","Penpot MCP: connect, verify, and fix streaming via nginx.","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fui-ux\u002Fpenpot-mcp\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":86,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M7AYYNKP61CN2281VMJDXMB5","nginx-streaming-proxy","Configure nginx for streaming endpoints (SSE, MCP, WS).","Unlicense","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fui-ux\u002Fnginx-streaming-proxy\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M7AYYJP0MDV1D3WVEEZB3SGA","branding-ci","Use for brand\u002FCI identity, style guides, and brand docs.","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fui-ux\u002Fbranding-ci\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":86,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":19},"01M7AYYGCXVT97DDTCM0ZJT6F6","yuanbao","Yuanbao (元宝) groups: @mention users, query info\u002Fmembers.","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fshared-skills\u002Fyuanbao\u002FSKILL.md",[],[],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":86,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":19},"01M7AYYC2Y65NMGMD4GQXAZMMS","slidev","Create and present web-based slidedecks for developers using Slidev with Markdown, Vue components, code highlighting, animations, and interactive features. Use when building technical presentations, conference talks, code walkthroughs, teaching materials, or developer decks.","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fshared-skills\u002Fslidev\u002FSKILL.md",[],[]]