[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"extension-skill-web-infra-dev-browser-automation-ko":3,"guides-for-web-infra-dev-browser-automation":221,"similar-k17c1ke31fxn3ysjew1bygd90s867597":222},{"_creationTime":4,"_id":5,"children":6,"community":7,"display":9,"evaluation":21,"identity":186,"isFallback":191,"parentExtension":192,"providers":193,"relations":198,"repo":200,"workflow":218},1778054040814.547,"k17c1ke31fxn3ysjew1bygd90s867597",[],{"reviewCount":8},0,{"description":10,"installMethods":11,"name":12,"sourceUrl":13,"tags":14},"Vision-driven browser automation using Midscene. Operates from screenshots — no DOM or accessibility labels needed. Runs in headless Puppeteer — does NOT take over the user's mouse or keyboard. Also supports CDP mode and Bridge mode to connect to an existing Chrome. Use this skill when the user wants to: - Browse, navigate, or open web pages - Scrape, extract, or collect data from websites - Fill out forms, click buttons, or interact with web elements - Verify, validate, test, or QA frontend UI behavior - Take screenshots of web pages - Automate multi-step web workflows - Test what was just built, see if it works in browser - Connect to Chrome via CDP, DevTools Protocol, or remote debugging - Connect to user's Chrome browser, control my browser, operate my Chrome Powered by Midscene.js (https://midscenejs.com)",{},"Midscene Browser Automation","https://github.com/web-infra-dev/midscene-skills/tree/HEAD/skills/browser",[15,16,17,18,19,20],"browser","automation","puppeteer","cdp","web-scraping","ui-automation",{"_creationTime":22,"_id":23,"extensionId":5,"locale":24,"result":25,"trustSignals":175,"workflow":184},1778054071592.8618,"kn7efsmec323zqhghranh6fw9h866hqe","en",{"checks":26,"evaluatedAt":165,"extensionSummary":166,"promptVersionExtension":167,"promptVersionScoring":168,"rationale":169,"score":170,"summary":171,"tags":172,"targetMarket":173,"tier":174},[27,32,35,38,42,45,49,53,56,59,63,68,71,76,79,82,85,88,91,94,97,101,105,110,114,117,120,123,127,130,133,136,139,142,146,149,152,155,158,162],{"category":28,"check":29,"severity":30,"summary":31},"Practical Utility","Problem relevance","pass","The description clearly states the problem of vision-driven browser automation and lists specific use cases like browsing, scraping, and form filling.",{"category":28,"check":33,"severity":30,"summary":34},"Unique selling proposition","The skill offers significant value by operating from screenshots, enabling cross-platform reliability and avoiding DOM/accessibility label dependencies, which is a unique approach compared to standard browser automation.",{"category":28,"check":36,"severity":30,"summary":37},"Production readiness","The skill provides a complete lifecycle for browser automation, including connection, interaction, assertion, and reporting, with clear instructions for different modes and prerequisites.",{"category":39,"check":40,"severity":30,"summary":41},"Scope","Single responsibility principle","The skill focuses on browser automation and related interactions, maintaining a clear scope without branching into unrelated domains.",{"category":39,"check":43,"severity":30,"summary":44},"Description quality","The description is accurate, concise, readable, and well-structured, accurately reflecting the skill's capabilities and use cases.",{"category":46,"check":47,"severity":30,"summary":48},"Invocation","Scoped tools","The tools provided (connect, act, take_screenshot, assert, tap, disconnect, close, report-tool) are specific verb-noun actions within the browser automation domain.",{"category":50,"check":51,"severity":30,"summary":52},"Documentation","Configuration & parameter reference","All necessary configuration variables (MIDSCENE_MODEL_*) and command-line parameters for different modes (CDP, Bridge) are documented with clear examples.",{"category":39,"check":54,"severity":30,"summary":55},"Tool naming","Tool names like `take_screenshot`, `connect`, `act`, and `assert` are descriptive verb-noun pairs within the browser automation domain.",{"category":39,"check":57,"severity":30,"summary":58},"Minimal I/O surface","Inputs for commands like `act` and `assert` take natural language prompts, and outputs are generally focused on task completion or reporting, not excessive diagnostics.",{"category":60,"check":61,"severity":30,"summary":62},"License","License usability","The extension is licensed under the MIT License, a permissive open-source license.",{"category":64,"check":65,"severity":66,"summary":67},"Maintenance","Commit recency","not_applicable","The extension is evaluated as a part of a larger repository, and there is no specific commit date available for this individual skill.",{"category":64,"check":69,"severity":66,"summary":70},"Dependency Management","No third-party dependencies are explicitly managed within this skill's files; the installation instructions suggest using npm/pnpm, implying dependency management occurs at that level.",{"category":72,"check":73,"severity":74,"summary":75},"Security","Secret Management","info","The documentation mentions environment variables for API keys, which are standard practice for secret management. It advises against committing secrets to files. No specific secrets are hardcoded in the provided skill files.",{"category":72,"check":77,"severity":30,"summary":78},"Injection","The skill uses explicit tools and prompts for interactions rather than executing arbitrary commands or data loaded from untrusted sources.",{"category":72,"check":80,"severity":30,"summary":81},"Transitive Supply-Chain Grenades","The skill uses `npx` to execute commands from a known package (`@midscene/web`) and does not appear to fetch or execute arbitrary remote code. Dependencies are managed via standard package managers.",{"category":72,"check":83,"severity":30,"summary":84},"Sandbox Isolation","The skill operates through the `@midscene/web` tool, which is expected to have its own sandboxing. The skill's own scripts do not attempt to access files outside the intended scope.",{"category":72,"check":86,"severity":30,"summary":87},"Sandbox escape primitives","No detached process spawns or retry loops around denied tool calls were found in the skill's SKILL.md.",{"category":72,"check":89,"severity":30,"summary":90},"Data Exfiltration","The skill focuses on browser automation and does not appear to have any mechanisms for exfiltrating confidential data or credentials.",{"category":72,"check":92,"severity":30,"summary":93},"Hidden Text Tricks","Bundled files were inspected and found to be free of hidden-steering tricks, invisible Unicode characters, or other obfuscation methods.",{"category":72,"check":95,"severity":30,"summary":96},"Opaque code execution","The skill's instructions are written in plain markdown and do not involve obfuscated code, base64 payloads, or runtime script fetching.",{"category":98,"check":99,"severity":30,"summary":100},"Portability","Structural Assumption","The skill relies on the `@midscene/web` tool and standard browser debugging protocols, making no assumptions about specific user project directory structures.",{"category":102,"check":103,"severity":66,"summary":104},"Trust","Issues Attention","Issue data is not available for this specific skill.",{"category":106,"check":107,"severity":108,"summary":109},"Versioning","Release Management","warning","No explicit version information (e.g., version field in frontmatter, CHANGELOG) is present for this skill, and installation instructions might point to the 'main' branch.",{"category":111,"check":112,"severity":30,"summary":113},"Code Execution","Validation","The skill relies on the `@midscene/web` tool's internal validation for commands and parameters, and prompts are handled by the LLM with the expectation of structured interaction.",{"category":72,"check":115,"severity":30,"summary":116},"Unguarded Destructive Operations","The skill's operations are primarily focused on browser interaction and data retrieval, not destructive file system or infrastructure changes that would require additional guards.",{"category":111,"check":118,"severity":30,"summary":119},"Error Handling","The skill's instructions guide the LLM to handle errors gracefully by reporting results and troubleshooting connection issues, implying reliance on the underlying `@midscene/web` tool's error handling.",{"category":111,"check":121,"severity":66,"summary":122},"Logging","No explicit local audit file logging is implemented within the skill itself; the user is expected to review command outputs and potentially generated reports.",{"category":124,"check":125,"severity":74,"summary":126},"Compliance","GDPR","The skill operates on web pages, which may contain personal data. While the skill itself doesn't submit data to third parties, the underlying LLM processing and browser interactions could potentially expose personal data if not handled carefully by the user or the model.",{"category":124,"check":128,"severity":30,"summary":129},"Target market","The extension targets browser automation, which is a global functionality. No specific regional limitations or jurisdiction-specific logic were detected.",{"category":98,"check":131,"severity":30,"summary":132},"Runtime stability","The skill relies on standard Node.js environments and browser automation protocols, with instructions for handling CDP/Bridge modes, suggesting good cross-platform compatibility.",{"category":46,"check":134,"severity":30,"summary":135},"Precise Purpose","The skill clearly defines its purpose as vision-driven browser automation and provides specific use cases, boundaries, and triggers.",{"category":46,"check":137,"severity":30,"summary":138},"Concise Frontmatter","The frontmatter is concise and clearly summarizes the core capability and use cases for precise routing.",{"category":50,"check":140,"severity":30,"summary":141},"Concise Body","The SKILL.md body is well-structured and reasonably concise, delegating deeper material to examples and external documentation links.",{"category":143,"check":144,"severity":30,"summary":145},"Context","Progressive Disclosure","The skill uses relative paths to link to external documentation (e.g., Midscene.js website) and organizes commands clearly, following progressive disclosure principles.",{"category":143,"check":147,"severity":66,"summary":148},"Forked exploration","The skill's primary function is direct browser automation, not deep code review or extensive exploration that would necessitate `context: fork`.",{"category":28,"check":150,"severity":30,"summary":151},"Usage examples","Sufficient end-to-end examples are provided for various commands and modes, demonstrating input, invocation, and expected outcomes.",{"category":28,"check":153,"severity":30,"summary":154},"Edge cases","The skill addresses potential issues like connection failures, API key errors, timeouts, and dependency updates with clear troubleshooting steps and recovery advice.",{"category":111,"check":156,"severity":30,"summary":157},"Tool Fallback","The skill explicitly mentions fallbacks to Puppeteer mode if CDP or Bridge modes are unavailable and provides instructions for setting up the necessary tools like Node.js and the Midscene CLI package.",{"category":159,"check":160,"severity":30,"summary":161},"Safety","Halt on unexpected state","The instructions emphasize synchronous command execution and reporting results, implying a fail-closed behavior on unexpected states, and the troubleshooting section addresses connection failures.",{"category":98,"check":163,"severity":30,"summary":164},"Cross-skill coupling","The skill is self-contained and focuses solely on browser automation, with no implicit reliance on other skills. Cross-skill coordination is not a primary concern for its scope.",1778054054975,"This skill leverages Midscene.js to control web browsers from screenshots, supporting headless Puppeteer, CDP, and Bridge modes. It allows for complex interactions like form filling, data scraping, and UI testing, with clear instructions for setup and usage.","2.0.0","3.4.0","The browser automation skill is well-documented, provides clear examples, and handles various connection modes and error conditions. The core functionality is unique and production-ready. The only minor concern is the lack of explicit versioning for the skill itself, though the underlying package might be versioned.",92,"A robust and well-documented vision-driven browser automation skill.",[15,16,17,18,19,20],"global","verified",{"codeQuality":176,"collectedAt":177,"documentation":178,"maintenance":180,"security":181,"testCoverage":183},{},1778054043064,{"descriptionLength":179,"readmeSize":8},821,{},{"hasNpmPackage":182,"smitheryVerified":182},false,{"hasCi":182,"hasTests":182},{"updatedAt":185},1778054071592,{"githubOwner":187,"githubRepo":188,"locale":24,"slug":189,"type":190},"web-infra-dev","midscene-skills","browser-automation","skill",true,null,{"extract":194,"llm":197},{"commitSha":195,"license":196},"f36878ba95af82e4080e73b625d61a7df77b1aeb","MIT",{"promptVersionExtension":167,"promptVersionScoring":168,"score":170,"targetMarket":173,"tier":174},{"repoId":199},"kd735e4mxzrfmjn5ja3fyx62fs865b72",{"_creationTime":201,"_id":199,"identity":202,"providers":204,"workflow":215},1777995558409.8755,{"githubOwner":187,"githubRepo":188,"sourceUrl":203},"https://github.com/web-infra-dev/midscene-skills",{"discover":205,"github":208},{"sources":206},[207],"skills-sh",{"closedIssues90d":8,"forks":209,"homepage":210,"license":196,"openIssues90d":8,"pushedAt":211,"readmeSize":212,"stars":213,"topics":214},15,"https://midscenejs.com/skills.html",1777460604000,3845,205,[],{"discoverAt":216,"extractAt":217,"githubAt":217,"updatedAt":217},1777995558409,1778054042066,{"anyEnrichmentAt":219,"extractAt":220,"githubAt":219,"llmAt":185,"updatedAt":185},1778054041174,1778054040814,[],[223,248,278,303,330,359],{"_creationTime":224,"_id":225,"community":226,"display":227,"identity":234,"providers":237,"relations":241,"workflow":244},1778053197391.3772,"k17ewq1gx45xdgk7ahghe8demx866j56",{"reviewCount":8},{"description":228,"installMethods":229,"name":230,"sourceUrl":231,"tags":232},"Minimal Chrome DevTools Protocol tools for browser automation and scraping. Use when you need to start Chrome, navigate pages, execute JavaScript, take screenshots, or interactively pick DOM elements. Triggers include \"browse website\", \"scrape page\", \"take screenshot\", \"automate browser\", \"extract DOM\", \"web scraping\".",{},"Browser Tools","https://github.com/iamzhihuix/happy-claude-skills/tree/HEAD/skills/browser",[15,16,233,18,17],"scraping",{"githubOwner":235,"githubRepo":236,"locale":24,"slug":15,"type":190},"iamzhihuix","happy-claude-skills",{"extract":238,"llm":240},{"commitSha":239,"license":196},"f49e7782a551759c9f9e0a4d4417ff053f0a86fd",{"promptVersionExtension":167,"promptVersionScoring":168,"score":170,"targetMarket":173,"tier":174},{"parentExtensionId":242,"repoId":243},"k17bs2gsezsesteta4vpdypgen866fae","kd7dbbtdq95nkcs3k7fg9w6fdn864j0b",{"anyEnrichmentAt":245,"extractAt":246,"githubAt":245,"llmAt":247,"updatedAt":247},1778053199195,1778053197391,1778053284450,{"_creationTime":249,"_id":250,"community":251,"display":252,"identity":263,"providers":266,"relations":271,"workflow":274},1778054268187.7783,"k1799kwx7k8g1vx165qr4np3298670sw",{"reviewCount":8},{"description":253,"installMethods":254,"name":255,"sourceUrl":256,"tags":257},"Guide for using the Bright Data CLI (`brightdata` / `bdata`) to scrape websites, search the web, extract structured data from 40+ platforms, manage proxy zones, and check account budget. Use this skill whenever the user wants to scrape a URL, search Google/Bing/Yandex, extract data from Amazon/LinkedIn/Instagram/TikTok/YouTube/Reddit or any other platform, check their Bright Data balance or zones, or do anything involving web data collection from the terminal. Also trigger when the user mentions brightdata, bdata, web scraping CLI, SERP API, or wants to install Bright Data skills into their coding agent.",{},"Bright Data CLI","https://github.com/brightdata/skills/tree/HEAD/skills/brightdata-cli",[258,259,19,260,261,16,262],"brightdata","cli","data-extraction","serp-api","terminal",{"githubOwner":258,"githubRepo":264,"locale":24,"slug":265,"type":190},"skills","brightdata-cli",{"extract":267,"llm":269},{"commitSha":268,"license":196},"d0eeb1fbab809ffffe7c270186bd3eb78cf0c8ba",{"promptVersionExtension":167,"promptVersionScoring":168,"score":270,"targetMarket":173,"tier":174},99,{"parentExtensionId":272,"repoId":273},"k177secs2fy2665c3z8prspg0s867xd1","kd7e4q3ah25vmt87x67vanphhn864r9h",{"anyEnrichmentAt":275,"extractAt":276,"githubAt":275,"llmAt":277,"updatedAt":277},1778054269540,1778054268187,1778054318963,{"_creationTime":279,"_id":280,"community":281,"display":282,"identity":290,"providers":292,"relations":297,"workflow":299},1778053148350.4246,"k17ff5jeywyjm6a5nrmcqtq199867shs",{"reviewCount":8},{"description":283,"installMethods":284,"name":285,"sourceUrl":286,"tags":287},"Automate web browser interactions, scraping, testing, and workflow automation with Puppeteer/Playwright",{},"Browser Automation","https://github.com/claude-office-skills/skills/tree/HEAD/browser-automation",[15,17,288,233,289,16],"playwright","testing",{"githubOwner":291,"githubRepo":264,"locale":24,"slug":189,"type":190},"claude-office-skills",{"extract":293,"llm":295},{"commitSha":294,"license":196},"9c4c7d5cd2813a8936bf2c9fdb174ea883b85a11",{"promptVersionExtension":167,"promptVersionScoring":168,"score":296,"targetMarket":173,"tier":174},95,{"repoId":298},"kd7fw7xbj58qc2z8whrrjptbed8659db",{"anyEnrichmentAt":300,"extractAt":301,"githubAt":300,"llmAt":302,"updatedAt":302},1778053151766,1778053148350,1778053561145,{"_creationTime":304,"_id":305,"community":306,"display":307,"identity":315,"providers":319,"relations":324,"workflow":326},1778053380851.3115,"k178z7154hzyj4f26ptbfj7zh5866p5z",{"reviewCount":8},{"description":308,"installMethods":309,"name":310,"sourceUrl":311,"tags":312},"Headless browser automation, CDP discovery, and cross-browser cookie extraction. Internal building block for the `react-doctor browser ...` CLI.",{},"Headless Browser","https://github.com/millionco/react-doctor/tree/HEAD/packages/headless-browser",[15,16,18,313,314,288],"cookies","headless",{"githubOwner":316,"githubRepo":317,"locale":24,"slug":318,"type":190},"millionco","react-doctor","react-doctor-browser",{"extract":320,"llm":322},{"commitSha":321,"license":196},"0053faa0a85dda7d57bc09764ab600655829a4c0",{"promptVersionExtension":167,"promptVersionScoring":168,"score":323,"targetMarket":173,"tier":174},90,{"repoId":325},"kd7c7y7mgbqtgjw8q3h5pevtgh864s4d",{"anyEnrichmentAt":327,"extractAt":328,"githubAt":327,"llmAt":329,"updatedAt":329},1778053381509,1778053380851,1778053395404,{"_creationTime":331,"_id":332,"community":333,"display":334,"identity":345,"providers":349,"relations":353,"workflow":355},1778053622473.6482,"k177qzhg585h8jr95mrhazbca9867398",{"reviewCount":8},{"description":335,"installMethods":336,"name":337,"sourceUrl":338,"tags":339},"Create and manipulate PowerPoint PPTX files programmatically. Use when the user needs to generate presentations, modify PPTX templates, extract slide content, create thumbnail previews, or automate PowerPoint workflows. Supports both template-based generation (for branding compliance) and from-scratch creation. Keywords: PowerPoint, PPTX, presentation, slides, template, deck, slideshow, corporate, branding.",{},"PPTX Generator","https://github.com/jwynia/agent-skills/tree/HEAD/skills/general/document-processing/presentation/pptx-generator",[340,341,342,343,16,344],"powerpoint","pptx","presentation","generator","document-processing",{"githubOwner":346,"githubRepo":347,"locale":24,"slug":348,"type":190},"jwynia","agent-skills","pptx-generator",{"extract":350,"llm":352},{"commitSha":351,"license":196},"e02ec7e226a6e4f8419fd3b88a1d8e472d421b32",{"promptVersionExtension":167,"promptVersionScoring":168,"score":270,"targetMarket":173,"tier":174},{"repoId":354},"kd7efn3mprpa8rd8vm5hw5ebzx864fph",{"anyEnrichmentAt":356,"extractAt":357,"githubAt":356,"llmAt":358,"updatedAt":358},1778053625386,1778053622473,1778054012696,{"_creationTime":360,"_id":361,"community":362,"display":363,"identity":373,"providers":377,"relations":382,"workflow":384},1778054781976.5928,"k1779jg9bachbejc4hfmyejtjx867tzc",{"reviewCount":8},{"description":364,"installMethods":365,"name":366,"sourceUrl":367,"tags":368},"AI Native Camp Day 2 Context Sync 스킬 만들기. 여러 외부 도구에서 컨텍스트를 수집하여 하나의 sync 문서로 만드는 나만의 스킬을 직접 구축한다. \"2일차\", \"Day 2\", \"context sync\", \"컨텍스트 싱크\", \"sync 스킬\", \"스킬 만들기\", \"정보 수집 스킬\" 요청에 사용.",{},"Day 2: 나만의 Context Sync 스킬 만들기","https://github.com/ai-native-camp/camp-1/tree/HEAD/.agents/skills/day2-create-context-sync-skill",[16,369,370,371,372],"mcp","skill-building","productivity","data-sync",{"githubOwner":374,"githubRepo":375,"locale":24,"slug":376,"type":190},"ai-native-camp","camp-1","day2-create-context-sync-skill",{"extract":378,"llm":380},{"commitSha":379,"license":66},"9ffaf358dc8c88567d8f0450966b5518071da4f0",{"promptVersionExtension":167,"promptVersionScoring":168,"score":381,"targetMarket":173,"tier":174},98,{"repoId":383},"kd72seepns71xx9ksxrb02bs1n8645k6",{"anyEnrichmentAt":385,"extractAt":386,"githubAt":385,"llmAt":387,"updatedAt":387},1778054782298,1778054781976,1778054817045]