[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"extension-plugin-browserbase-skills-de":3,"guides-for-browserbase-skills":784,"similar-k17bx77jb71yrwatep2nb43r0d86m6cb-de":785},{"_creationTime":4,"_id":5,"children":6,"community":284,"display":285,"evaluation":289,"identity":516,"isFallback":500,"parentExtension":519,"providers":542,"relations":548,"repo":549,"tags":782,"workflow":783},1778683460321.38,"k17bx77jb71yrwatep2nb43r0d86m6cb",[7,47,70,95,116,138,160,180,202,224,243,263],{"_creationTime":8,"_id":9,"community":10,"display":12,"identity":18,"providers":25,"relations":41,"tags":43,"workflow":44},1778683460321.381,"k1707rtzme0qezretvp3gmtfvn86n1k1",{"reviewCount":11},0,{"description":13,"installMethods":14,"name":16,"sourceUrl":17},"Self-improving browser automation via the auto-research loop. Iteratively runs a browsing task, reads the trace, and improves the navigation skill (strategy.md) until it reliably passes. Supports parallel runs across multiple tasks using sub-agents. Use when you want to build or improve browser automation skills for specific website tasks.",{"claudeCode":15},"browserbase/skills","AutoBrowse","https://github.com/browserbase/skills",{"basePath":19,"githubOwner":20,"githubRepo":21,"locale":22,"slug":23,"type":24},"skills/autobrowse","browserbase","skills","en","autobrowse","skill",{"evaluate":26,"extract":38},{"promptVersionExtension":27,"promptVersionScoring":28,"score":29,"tags":30,"targetMarket":36,"tier":37},"3.0.0","4.4.0",100,[31,32,33,34,35],"browser-automation","web-scraping","llm-agent","testing","node-js","global","verified",{"commitSha":39,"license":40},"HEAD","MIT",{"parentExtensionId":5,"repoId":42},"kd77wvcdm5fq9xp8hk6ppm832s86myxr",[31,33,35,34,32],{"evaluatedAt":45,"extractAt":46,"updatedAt":45},1778683610531,1778683460321,{"_creationTime":48,"_id":49,"community":50,"display":51,"identity":55,"providers":57,"relations":66,"tags":67,"workflow":68},1778683460321.3813,"k175dd0cx0atq57d86t68p0ftn86mx3n",{"reviewCount":11},{"description":52,"installMethods":53,"name":54,"sourceUrl":17},"Automate web browser interactions using natural language via CLI commands. Use when the user asks to browse websites, navigate web pages, extract data from websites, take screenshots, fill forms, click buttons, or interact with web applications. Supports remote Browserbase sessions with automatic CAPTCHA solving, anti-bot stealth mode, and residential proxies — ideal for scraping protected websites, bypassing bot detection, and interacting with JavaScript-heavy pages.",{"claudeCode":15},"browser",{"basePath":56,"githubOwner":20,"githubRepo":21,"locale":22,"slug":54,"type":24},"skills/browser",{"evaluate":58,"extract":65},{"promptVersionExtension":27,"promptVersionScoring":28,"score":59,"tags":60,"targetMarket":36,"tier":37},95,[31,32,61,62,63,64],"cli","web-interaction","remote-sessions","captcha-solving",{"commitSha":39},{"parentExtensionId":5,"repoId":42},[31,64,61,63,62,32],{"evaluatedAt":69,"extractAt":46,"updatedAt":69},1778683633116,{"_creationTime":71,"_id":72,"community":73,"display":74,"identity":78,"providers":81,"relations":91,"tags":92,"workflow":93},1778683460321.3816,"k17fd1dhmc76frn5d1xjjc23bx86mjba",{"reviewCount":11},{"description":75,"installMethods":76,"name":77,"sourceUrl":17},"Capture a full DevTools-protocol trace of any browser automation — CDP firehose, screenshots, and DOM dumps — then bisect the stream into per-page searchable buckets. Use when the user wants to debug a failed run, audit network/console/DOM activity, attach a trace to an in-progress session, or feed structured per-page summaries back into an agent loop so its next iteration learns from the last one.",{"claudeCode":15},"Browser Trace",{"basePath":79,"githubOwner":20,"githubRepo":21,"locale":22,"slug":80,"type":24},"skills/browser-trace","browser-trace",{"evaluate":82,"extract":90},{"promptVersionExtension":27,"promptVersionScoring":28,"score":83,"tags":84,"targetMarket":36,"tier":37},99,[31,85,86,87,88,89],"debugging","tracing","cdp","developer-tools","network-analysis",{"commitSha":39,"license":40},{"parentExtensionId":5,"repoId":42},[31,87,85,88,89,86],{"evaluatedAt":94,"extractAt":46,"updatedAt":94},1778683657573,{"_creationTime":96,"_id":97,"community":98,"display":99,"identity":103,"providers":105,"relations":112,"tags":113,"workflow":114},1778683460321.3818,"k177bgbdgskxdrd6hrsfrek44186m1xs",{"reviewCount":11},{"description":100,"installMethods":101,"name":102,"sourceUrl":17},"Use the Browserbase CLI (`bb`) for Browserbase Functions and platform API workflows. Use when the user asks to run `bb`, deploy or invoke functions, manage sessions, projects, contexts, or extensions, fetch a page through the Browserbase Fetch API, search the web through the Browserbase Search API, or scaffold starter templates. Prefer the Browser skill for interactive browsing; use `bb browse` only when the user explicitly wants the Browserbase CLI path.",{"claudeCode":15},"browserbase-cli",{"basePath":104,"githubOwner":20,"githubRepo":21,"locale":22,"slug":102,"type":24},"skills/browserbase-cli",{"evaluate":106,"extract":111},{"promptVersionExtension":27,"promptVersionScoring":28,"score":83,"tags":107,"targetMarket":36,"tier":37},[20,61,108,109,110],"automation","api","functions",{"commitSha":39},{"parentExtensionId":5,"repoId":42},[109,108,20,61,110],{"evaluatedAt":115,"extractAt":46,"updatedAt":115},1778683681669,{"_creationTime":117,"_id":118,"community":119,"display":120,"identity":124,"providers":127,"relations":134,"tags":135,"workflow":136},1778683460321.382,"k173t0vany5k5ve3tam3nyk8c186nq41",{"reviewCount":11},{"description":121,"installMethods":122,"name":123,"sourceUrl":17},"Company discovery and deep research skill. Researches a company's product and ICP,\ndiscovers target companies to sell to using Browserbase Search API, deeply researches\neach using a Plan→Research→Synthesize pattern, and scores ICP fit — compiled into\na scored research report and CSV. Supports depth modes (quick/deep/deeper) for\nbalancing scale vs intelligence.\nUse when the user wants to: (1) find companies to sell to, (2) research potential\ncustomers, (3) discover companies matching an ICP, (4) build a target company list,\n(5) do market research on prospects. Triggers: \"find companies to sell to\",\n\"company research\", \"find prospects\", \"ICP research\", \"target companies\",\n\"who should we sell to\", \"market research\", \"lead research\", \"prospect list\".\n",{"claudeCode":15},"Company Research",{"basePath":125,"githubOwner":20,"githubRepo":21,"locale":22,"slug":126,"type":24},"skills/company-research","company-research",{"evaluate":128,"extract":133},{"promptVersionExtension":27,"promptVersionScoring":28,"score":129,"tags":130,"targetMarket":36,"tier":37},98,[126,131,132,32,31],"lead-generation","sales-intelligence",{"commitSha":39,"license":40},{"parentExtensionId":5,"repoId":42},[31,126,131,132,32],{"evaluatedAt":137,"extractAt":46,"updatedAt":137},1778683708926,{"_creationTime":139,"_id":140,"community":141,"display":142,"identity":146,"providers":148,"relations":156,"tags":157,"workflow":158},1778683460321.3823,"k170yerym85rjxnc0dhv8ncq4186n1n4",{"reviewCount":11},{"description":143,"installMethods":144,"name":145,"sourceUrl":17},"Sync cookies from local Chrome to a Browserbase persistent context so the browse CLI can access authenticated sites. Use when the user wants to browse as themselves, sync cookies, or log into sites via Browserbase.",{"claudeCode":15},"cookie-sync",{"basePath":147,"githubOwner":20,"githubRepo":21,"locale":22,"slug":145,"type":24},"skills/cookie-sync",{"evaluate":149,"extract":155},{"promptVersionExtension":27,"promptVersionScoring":28,"score":150,"tags":151,"targetMarket":36,"tier":154},78,[31,152,153,20,61],"cookies","chrome","community",{"commitSha":39},{"parentExtensionId":5,"repoId":42},[31,20,153,61,152],{"evaluatedAt":159,"extractAt":46,"updatedAt":159},1778683727786,{"_creationTime":161,"_id":162,"community":163,"display":164,"identity":168,"providers":170,"relations":176,"tags":177,"workflow":178},1778683460321.3826,"k1749brrb1xg8wbt6802r76mr986mzcf",{"reviewCount":11},{"description":165,"installMethods":166,"name":167,"sourceUrl":17},"Event prospecting skill. Takes a conference / event speakers URL,\nextracts the people, filters their companies against the user's\nICP, then deep-researches only the speakers at ICP-fit companies.\nOutputs a person-first HTML report where each card answers \"why\nshould the AE talk to this person?\" with all public links and a\none-click DM opener.\nUse when the user wants to: (1) find leads at a specific\nconference, (2) prep for an event, (3) research event speakers,\n(4) build a target list from a sponsor/exhibitor page,\n(5) scrape conference speakers and rank by ICP fit.\nTriggers: \"find leads at {event}\", \"research speakers at\",\n\"prospect this conference\", \"stripe sessions leads\",\n\"ai engineer summit prospects\", \"event prospecting\",\n\"scrape conference speakers\", \"who should I meet at\".\n",{"claudeCode":15},"event-prospecting",{"basePath":169,"githubOwner":20,"githubRepo":21,"locale":22,"slug":167,"type":24},"skills/event-prospecting",{"evaluate":171,"extract":175},{"promptVersionExtension":27,"promptVersionScoring":28,"score":129,"tags":172,"targetMarket":36,"tier":37},[167,131,32,173,174],"data-extraction","reporting",{"commitSha":39},{"parentExtensionId":5,"repoId":42},[173,167,131,174,32],{"evaluatedAt":179,"extractAt":46,"updatedAt":179},1778683760146,{"_creationTime":181,"_id":182,"community":183,"display":184,"identity":188,"providers":190,"relations":198,"tags":199,"workflow":200},1778683460321.3828,"k178byhs7vswk8154tmztnehgx86n47n",{"reviewCount":11},{"description":185,"installMethods":186,"name":187,"sourceUrl":17},"Use this skill when the user wants to retrieve a URL without a full browser session: fetch HTML or JSON from static pages, inspect status codes or headers, follow redirects, or get page source for simple scraping. Prefer it over a browser when JavaScript rendering and page interaction are not needed. Supports proxies and redirect control.",{"claudeCode":15},"fetch",{"basePath":189,"githubOwner":20,"githubRepo":21,"locale":22,"slug":187,"type":24},"skills/fetch",{"evaluate":191,"extract":197},{"promptVersionExtension":27,"promptVersionScoring":28,"score":129,"tags":192,"targetMarket":36,"tier":37},[193,194,195,109,196],"web","http","scraping","data-retrieval",{"commitSha":39},{"parentExtensionId":5,"repoId":42},[109,196,194,195,193],{"evaluatedAt":201,"extractAt":46,"updatedAt":201},1778683789383,{"_creationTime":203,"_id":204,"community":205,"display":206,"identity":209,"providers":211,"relations":220,"tags":221,"workflow":222},1778683460321.383,"k17f1z6rytr3pcs0qb838kebv586mhyh",{"reviewCount":11},{"description":207,"installMethods":208,"name":110,"sourceUrl":17},"Deploy serverless browser automation as cloud functions using Browserbase. Use when the user wants to deploy browser automation to run on a schedule or cron, create a webhook endpoint for browser tasks, run automation in the cloud instead of locally, or asks about Browserbase Functions.",{"claudeCode":15},{"basePath":210,"githubOwner":20,"githubRepo":21,"locale":22,"slug":110,"type":24},"skills/functions",{"evaluate":212,"extract":219},{"promptVersionExtension":27,"promptVersionScoring":28,"score":213,"tags":214,"targetMarket":36,"tier":154},79,[31,215,216,61,217,218],"cloud-functions","serverless","typescript","playwright",{"commitSha":39},{"parentExtensionId":5,"repoId":42},[31,61,215,218,216,217],{"evaluatedAt":223,"extractAt":46,"updatedAt":223},1778683808928,{"_creationTime":225,"_id":226,"community":227,"display":228,"identity":232,"providers":234,"relations":239,"tags":240,"workflow":241},1778683460321.3833,"k1778nak5hc8b0f1v34mcmqsxd86mh6n",{"reviewCount":11},{"description":229,"installMethods":230,"name":231,"sourceUrl":17},"Build local constrained-browser agents with a safe_browser tool that owns CDP, enforces a domain allowlist with Fetch interception, and lets a runtime Claude Agent SDK agent complete browsing tasks without raw browser, shell, or CDP access. Use when the user wants an agent to browse or scrape while staying on approved domains, demo blocked off-domain navigation, or generate a safe browser client.",{"claudeCode":15},"safe-browser",{"basePath":233,"githubOwner":20,"githubRepo":21,"locale":22,"slug":231,"type":24},"skills/safe-browser",{"evaluate":235,"extract":238},{"promptVersionExtension":27,"promptVersionScoring":28,"score":83,"tags":236,"targetMarket":36,"tier":37},[54,237,195,108,87],"agent-sdk",{"commitSha":39,"license":40},{"parentExtensionId":5,"repoId":42},[237,108,54,87,195],{"evaluatedAt":242,"extractAt":46,"updatedAt":242},1778683843366,{"_creationTime":244,"_id":245,"community":246,"display":247,"identity":251,"providers":253,"relations":259,"tags":260,"workflow":261},1778683460321.3835,"k17dj26t9cgd67nx9b189zjaax86n8z6",{"reviewCount":11},{"description":248,"installMethods":249,"name":250,"sourceUrl":17},"Use this skill when the user wants to search the web without a full browser session: find URLs, titles, and metadata for a query. Prefer it over a browser when you just need search results, not page content. Returns structured results with titles, URLs, authors, and dates.",{"claudeCode":15},"search",{"basePath":252,"githubOwner":20,"githubRepo":21,"locale":22,"slug":250,"type":24},"skills/search",{"evaluate":254,"extract":258},{"promptVersionExtension":27,"promptVersionScoring":28,"score":255,"tags":256,"targetMarket":36,"tier":37},97,[250,193,109,108,257],"information-retrieval",{"commitSha":39},{"parentExtensionId":5,"repoId":42},[109,108,257,250,193],{"evaluatedAt":262,"extractAt":46,"updatedAt":262},1778683865815,{"_creationTime":264,"_id":265,"community":266,"display":267,"identity":271,"providers":273,"relations":280,"tags":281,"workflow":282},1778683460321.3838,"k17dn7zcf77stm2nh1701tbywh86nc09",{"reviewCount":11},{"description":268,"installMethods":269,"name":270,"sourceUrl":17},"AI-powered adversarial UI testing via the browse CLI. Analyzes git diffs to test only what changed, or explores the full app to find bugs. Tests functional correctness, accessibility, responsive layout, and UX heuristics. Use when the user asks to test UI changes, QA a pull request, audit accessibility, or run exploratory testing. Supports local browser (localhost) and remote Browserbase (deployed sites).",{"claudeCode":15},"ui-test",{"basePath":272,"githubOwner":20,"githubRepo":21,"locale":22,"slug":270,"type":24},"skills/ui-test",{"evaluate":274,"extract":279},{"promptVersionExtension":27,"promptVersionScoring":28,"score":255,"tags":275,"targetMarket":36,"tier":37},[276,31,277,278,34],"ui-testing","qa","accessibility",{"commitSha":39},{"parentExtensionId":5,"repoId":42},[278,31,277,34,276],{"evaluatedAt":283,"extractAt":46,"updatedAt":283},1778683891525,{"reviewCount":11},{"description":286,"installMethods":287,"name":288,"sourceUrl":17},"Automate web browser interactions using natural language. Use when the user asks to browse websites, navigate web pages, extract data from websites, take screenshots, fill forms, click buttons, or interact with web applications.",{"claudeCode":21},"browse",{"_creationTime":290,"_id":291,"extensionId":5,"locale":22,"result":292,"trustSignals":498,"workflow":514},1778683581384.1226,"kn788729q2y1jxdhsc7wxxdgqd86mjbs",{"checks":293,"evaluatedAt":476,"extensionSummary":477,"features":478,"nonGoals":484,"promptVersionExtension":27,"promptVersionScoring":28,"purpose":488,"rationale":489,"score":129,"summary":490,"tags":491,"targetMarket":36,"tier":37,"useCases":493},[294,299,302,305,309,312,316,321,324,327,331,335,338,342,345,348,351,354,357,360,364,368,372,376,380,383,386,389,393,396,399,402,405,408,411,415,419,423,426,430,433,436,439,442,444,447,450,453,455,458,462,465,468,472],{"category":295,"check":296,"severity":297,"summary":298},"Practical Utility","Problem relevance","pass","The description clearly states the problem of building local constrained-browser agents and the need for domain-specific browsing tasks without raw browser access.",{"category":295,"check":300,"severity":297,"summary":301},"Unique selling proposition","The safe_browser tool, CDP ownership, domain allowlist enforcement, and Fetch interception offer significant value beyond a standard browser tool.",{"category":295,"check":303,"severity":297,"summary":304},"Production readiness","The skill provides a well-defined tool and a clear example for building constrained browser agents, indicating readiness for integration into agent workflows.",{"category":306,"check":307,"severity":297,"summary":308},"Scope","Single responsibility principle","The skill focuses on creating a constrained browser agent with specific tools and domain enforcement, adhering to a single responsibility.",{"category":306,"check":310,"severity":297,"summary":311},"Description quality","The description accurately reflects the skill's functionality: building local agents with a safe_browser tool, CDP ownership, domain allowlisting, and Fetch interception.",{"category":313,"check":314,"severity":297,"summary":315},"Invocation","Scoped tools","The skill exposes a single, well-defined `safe_browser` tool that encapsulates CDP control and domain enforcement.",{"category":317,"check":318,"severity":319,"summary":320},"Documentation","Configuration & parameter reference","not_applicable","The skill does not expose configurable parameters or options that require documentation beyond its core functionality.",{"category":306,"check":322,"severity":297,"summary":323},"Tool naming","The single tool, `safe_browser`, is descriptively named and clearly indicates its function.",{"category":306,"check":325,"severity":297,"summary":326},"Minimal I/O surface","The `safe_browser` tool's operations are focused on browsing within allowlisted domains, with clear inputs and outputs implied by its purpose.",{"category":328,"check":329,"severity":297,"summary":330},"License","License usability","The extension is licensed under MIT, a permissive open-source license, with a LICENSE.txt file present.",{"category":332,"check":333,"severity":297,"summary":334},"Maintenance","Commit recency","The repository shows recent commits, indicating active maintenance.",{"category":332,"check":336,"severity":297,"summary":337},"Dependency Management","The skill has a lockfile (`package-lock.json`) and uses standard Node.js dependencies, indicating proper dependency management.",{"category":339,"check":340,"severity":319,"summary":341},"Security","Secret Management","The skill does not appear to handle or expose any secrets.",{"category":339,"check":343,"severity":297,"summary":344},"Injection","The skill uses Playwright/CDP and Fetch interception, which are designed to isolate and sanitize interactions, preventing injection through third-party data.",{"category":339,"check":346,"severity":297,"summary":347},"Transitive Supply-Chain Grenades","The skill relies on bundled dependencies (Playwright, Agent SDK) and does not fetch external code at runtime.",{"category":339,"check":349,"severity":297,"summary":350},"Sandbox Isolation","The skill operates within the agent's environment and uses Playwright/CDP for browser control, inherently isolating its operations.",{"category":339,"check":352,"severity":297,"summary":353},"Sandbox escape primitives","The skill's design focuses on controlled browser interaction via CDP and Fetch interception, with no apparent sandbox escape primitives.",{"category":339,"check":355,"severity":297,"summary":356},"Data Exfiltration","The skill is designed for controlled browsing within allowlisted domains and does not inherently exfiltrate confidential data.",{"category":339,"check":358,"severity":297,"summary":359},"Hidden Text Tricks","Bundled content is free of hidden-steering tricks, and descriptions are clean.",{"category":361,"check":362,"severity":297,"summary":363},"Hooks","Opaque code execution","The skill uses standard Node.js and Playwright/CDP, with no obfuscated or dynamically fetched code in its execution path.",{"category":365,"check":366,"severity":297,"summary":367},"Portability","Structural Assumption","The skill's assumptions are based on the agent SDK environment and Playwright, which are generally portable.",{"category":369,"check":370,"severity":297,"summary":371},"Trust","Issues Attention","The repository shows a low number of open issues and no closed issues in the last 90 days, with recent commits indicating maintainer activity.",{"category":373,"check":374,"severity":297,"summary":375},"Versioning","Release Management","The skill has a `metadata.version` field in its SKILL.md, indicating versioning is present.",{"category":377,"check":378,"severity":297,"summary":379},"Code Execution","Validation","The skill's design using the Agent SDK and Playwright for browser control implies validation of inputs and operations.",{"category":339,"check":381,"severity":297,"summary":382},"Unguarded Destructive Operations","The skill is designed for browsing and data extraction, not destructive operations.",{"category":377,"check":384,"severity":297,"summary":385},"Error Handling","The use of the Agent SDK and Playwright suggests robust error handling mechanisms are in place for browser operations.",{"category":377,"check":387,"severity":319,"summary":388},"Logging","The skill itself does not perform destructive actions or outbound calls that would necessitate local audit logging beyond what the Agent SDK provides.",{"category":390,"check":391,"severity":297,"summary":392},"Compliance","GDPR","The skill operates within the agent's defined environment and focuses on browser control, not directly processing personal data beyond what a user might browse.",{"category":390,"check":394,"severity":297,"summary":395},"Target market","The skill is designed for local browser agents and is not regionally scoped; targetMarket is global.",{"category":365,"check":397,"severity":297,"summary":398},"Runtime stability","The skill relies on Node.js and Playwright, which are cross-platform, and the example demonstrates local Chromium usage.",{"category":317,"check":400,"severity":297,"summary":401},"README","The README provides a clear description of the skill's purpose, use cases, and setup instructions.",{"category":306,"check":403,"severity":319,"summary":404},"Tool surface size","The skill exposes a single, primary tool (`safe_browser`), making it a single-tool extension.",{"category":313,"check":406,"severity":319,"summary":407},"Overlapping near-synonym tools","With only one core tool, there are no overlapping or near-synonym tools.",{"category":317,"check":409,"severity":297,"summary":410},"Phantom features","The README accurately describes the functionality of the `safe_browser` tool and its constraints.",{"category":412,"check":413,"severity":297,"summary":414},"Install","Installation instruction","The README provides clear installation instructions for Node.js, Playwright, and the Agent SDK template, along with example usage.",{"category":416,"check":417,"severity":297,"summary":418},"Errors","Actionable error messages","The example demonstrates how failures would be surfaced through the Agent SDK and Playwright, implying actionable error reporting.",{"category":420,"check":421,"severity":297,"summary":422},"Execution","Pinned dependencies","The project includes a `package-lock.json` file, ensuring pinned dependencies.",{"category":306,"check":424,"severity":319,"summary":425},"Dry-run preview","The skill's primary function is browser interaction and data extraction, not state-changing operations requiring a dry-run preview.",{"category":427,"check":428,"severity":297,"summary":429},"Protocol","Idempotent retry & timeouts","Browser operations are typically stateful and non-idempotent, but Playwright and CDP handle session management and errors, implying retry mechanisms.",{"category":390,"check":431,"severity":297,"summary":432},"Telemetry opt-in","The skill's example code does not indicate any telemetry collection; any such implementation would be opt-in via the Agent SDK.",{"category":313,"check":434,"severity":297,"summary":435},"Name collisions","The skill exposes a single, distinct tool (`safe_browser`) and is clearly named `safe-browser`, avoiding collisions.",{"category":313,"check":437,"severity":319,"summary":438},"Hooks-off mechanism","This skill does not utilize hooks in a way that requires a separate disable mechanism.",{"category":313,"check":440,"severity":319,"summary":441},"Hook matcher tightness","This skill does not define external hooks.",{"category":339,"check":443,"severity":319,"summary":441},"Hook security",{"category":361,"check":445,"severity":319,"summary":446},"Silent prompt rewriting","This skill does not define `UserPromptSubmit` hooks.",{"category":339,"check":448,"severity":319,"summary":449},"Permission Hook","This skill does not define `PermissionRequest` hooks.",{"category":390,"check":451,"severity":319,"summary":452},"Hook privacy","This skill does not define hooks that handle logging or telemetry.",{"category":377,"check":454,"severity":319,"summary":441},"Hook dependency",{"category":317,"check":456,"severity":297,"summary":457},"Feature Transparency","The README clearly explains the functionality, constraints, and example usage of the `safe_browser` tool.",{"category":459,"check":460,"severity":297,"summary":461},"Convention","Layout convention adherence","The skill follows standard Node.js project structure and provides a clear example within the template.",{"category":459,"check":463,"severity":319,"summary":464},"Plugin state","The skill itself is a builder, and its generated agents manage their own state within the agent's environment.",{"category":339,"check":466,"severity":319,"summary":467},"Keychain-stored secrets","The skill does not handle secrets directly; any secrets would be managed by the Agent SDK or user environment.",{"category":469,"check":470,"severity":297,"summary":471},"Dependencies","Tagged release sourcing","Dependencies like Playwright and the Agent SDK are managed via npm, implying tagged releases.",{"category":473,"check":474,"severity":319,"summary":475},"Installation","Clean uninstall","The skill is a builder and does not install background daemons or services that would prevent a clean uninstall.",1778683580242,"This skill provides the framework and example code to create local Claude Agent SDK applications that leverage a `safe_browser` tool. This tool manages CDP, enforces domain allowlists via Fetch interception, and limits the agent's browsing capabilities.",[479,480,481,482,483],"Creates constrained-browser agents","Manages CDP session","Enforces domain allowlist","Uses Fetch interception for requests","Provides a `safe_browser` tool",[485,486,487],"Providing raw browser, shell, or CDP access to the agent","Supporting off-domain navigation","Acting as a general-purpose browser automation tool outside the agent context","To enable developers to build secure browser-operating agents that strictly adhere to specified domains, enhancing safety and control in automated web interactions.","The skill is well-documented, secure, and adheres to best practices, with no critical or warning findings. Minor points like missing configuration reference and non-applicable checks contribute to a high score.","Builds constrained browser agents with a secure, domain-enforcing `safe_browser` tool.",[54,492,87,218,108],"agent",[494,495,496,497],"Building agents that must stay within approved websites","Demonstrating prompt-injection containment","Creating controlled web scrapers with domain policies","Generating browser agents as a security pattern",{"codeQuality":499,"collectedAt":501,"documentation":502,"maintenance":505,"popularity":511,"security":512,"testCoverage":513},{"hasLockfile":500},true,1778683557218,{"descriptionLength":503,"readmeSize":504},240,4101,{"closedIssues90d":11,"forks":506,"hasChangelog":507,"openIssues90d":508,"pushedAt":509,"stars":510},214,false,2,1778633729000,3179,{"npmDownloads":11},{"hasNpmPackage":500,"smitheryVerified":507},{"hasCi":507,"hasTests":507},{"updatedAt":515},1778683581384,{"basePath":517,"githubOwner":20,"githubRepo":21,"locale":22,"slug":21,"type":518},"","plugin",{"_creationTime":520,"_id":521,"community":522,"display":523,"identity":526,"parentExtension":528,"providers":529,"relations":538,"tags":539,"workflow":540},1778683460321.3796,"k17aawpn1wczwwhv1snb1x3q2h86m1hy",{"reviewCount":11},{"description":524,"installMethods":525,"name":20,"sourceUrl":17},"Browser automation and Browserbase CLI skills",{"claudeCode":15},{"basePath":517,"githubOwner":20,"githubRepo":21,"locale":22,"slug":21,"type":527},"marketplace",null,{"evaluate":530,"extract":533},{"promptVersionExtension":531,"promptVersionScoring":28,"score":59,"tags":532,"targetMarket":36,"tier":37},"3.1.0",[31,61,32,88,34],{"commitSha":39,"marketplace":534,"plugin":536},{"name":20,"pluginCount":535},1,{"mcpCount":11,"provider":537,"skillCount":11},"classify",{"repoId":42},[31,61,88,34,32],{"evaluatedAt":541,"extractAt":46,"updatedAt":541},1778683478405,{"evaluate":543,"extract":545},{"promptVersionExtension":27,"promptVersionScoring":28,"score":129,"tags":544,"targetMarket":36,"tier":37},[54,492,87,218,108],{"commitSha":39,"plugin":546},{"mcpCount":11,"provider":537,"skillCount":547},12,{"parentExtensionId":521,"repoId":42},{"_creationTime":550,"_id":42,"identity":551,"providers":552,"workflow":777},1778683454816.9644,{"githubOwner":20,"githubRepo":21,"sourceUrl":17},{"classify":553,"discover":768,"extract":771,"github":772,"npm":776},{"commitSha":39,"extensions":554},[555,565,570,575,580,585,590,612,620,644,651,672,682,702,710,717,729,737,760],{"basePath":517,"description":524,"displayName":20,"installMethods":556,"rationale":557,"selectedPaths":558,"source":564,"sourceLanguage":22,"type":527},{"claudeCode":15},"marketplace.json at .claude-plugin/marketplace.json",[559,562],{"path":560,"priority":561},".claude-plugin/marketplace.json","mandatory",{"path":563,"priority":561},"README.md","rule",{"basePath":517,"description":286,"displayName":288,"installMethods":566,"rationale":567,"selectedPaths":568,"source":564,"sourceLanguage":22,"type":518},{"claudeCode":21},"inline plugin source from marketplace.json at /",[569],{"path":563,"priority":561},{"basePath":517,"description":571,"displayName":110,"installMethods":572,"rationale":567,"selectedPaths":573,"source":564,"sourceLanguage":22,"type":518},"Deploy serverless browser automation to the cloud using Browserbase Functions. Use when the user wants to deploy automation to run on a schedule, needs a webhook endpoint for browser automation, wants to run automation in the cloud instead of locally, or asks about Browserbase Functions.",{"claudeCode":21},[574],{"path":563,"priority":561},{"basePath":517,"description":576,"displayName":102,"installMethods":577,"rationale":567,"selectedPaths":578,"source":564,"sourceLanguage":22,"type":518},"Use the Browserbase CLI (`bb`) for Browserbase Functions and platform API workflows. Use when the user asks to run `bb`, deploy or invoke functions, manage sessions, projects, contexts, or extensions, fetch a page through the Browserbase Fetch API, or open the Browserbase dashboard from the command line.",{"claudeCode":21},[579],{"path":563,"priority":561},{"basePath":517,"description":581,"displayName":80,"installMethods":582,"rationale":567,"selectedPaths":583,"source":564,"sourceLanguage":22,"type":518},"Capture a full DevTools-protocol trace (CDP firehose, screenshots, DOM dumps) alongside any browser automation, then bisect the stream into per-page searchable buckets. Use when the user wants to debug a failed run, audit network/console/DOM activity, attach a trace to an in-flight Browserbase session, or feed structured per-page summaries back into an agent loop.",{"claudeCode":21},[584],{"path":563,"priority":561},{"basePath":517,"description":586,"displayName":231,"installMethods":587,"rationale":567,"selectedPaths":588,"source":564,"sourceLanguage":22,"type":518},"Build local constrained-browser agents with a safe_browser tool that owns CDP, enforces a domain allowlist with Fetch interception, and lets a runtime Claude Agent SDK agent complete browsing tasks without raw browser, shell, or CDP access.",{"claudeCode":21},[589],{"path":563,"priority":561},{"basePath":19,"description":13,"displayName":23,"installMethods":591,"rationale":592,"selectedPaths":593,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/autobrowse/SKILL.md",[594,596,598,600,603,605,607,609],{"path":595,"priority":561},"SKILL.md",{"path":563,"priority":597},"high",{"path":599,"priority":597},"LICENSE.txt",{"path":601,"priority":602},"EXAMPLES.md","medium",{"path":604,"priority":602},"REFERENCE.md",{"path":606,"priority":602},"references/example-skill.md",{"path":608,"priority":602},"references/example-task.md",{"path":610,"priority":611},"scripts/evaluate.mjs","low",{"basePath":56,"description":52,"displayName":54,"installMethods":613,"rationale":614,"selectedPaths":615,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/browser/SKILL.md",[616,617,618,619],{"path":595,"priority":561},{"path":599,"priority":597},{"path":601,"priority":602},{"path":604,"priority":602},{"basePath":79,"description":75,"displayName":80,"installMethods":621,"rationale":622,"selectedPaths":623,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/browser-trace/SKILL.md",[624,625,626,627,628,630,632,634,636,638,640,642],{"path":595,"priority":561},{"path":599,"priority":597},{"path":601,"priority":602},{"path":604,"priority":602},{"path":629,"priority":611},"scripts/bb-capture.mjs",{"path":631,"priority":611},"scripts/bb-finalize.mjs",{"path":633,"priority":611},"scripts/bisect-cdp.mjs",{"path":635,"priority":611},"scripts/lib.mjs",{"path":637,"priority":611},"scripts/query.mjs",{"path":639,"priority":611},"scripts/snapshot-loop.mjs",{"path":641,"priority":611},"scripts/start-capture.mjs",{"path":643,"priority":611},"scripts/stop-capture.mjs",{"basePath":104,"description":100,"displayName":102,"installMethods":645,"rationale":646,"selectedPaths":647,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/browserbase-cli/SKILL.md",[648,649,650],{"path":595,"priority":561},{"path":599,"priority":597},{"path":604,"priority":602},{"basePath":125,"description":121,"displayName":126,"installMethods":652,"rationale":653,"selectedPaths":654,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/company-research/SKILL.md",[655,656,658,660,662,664,666,668,670],{"path":595,"priority":561},{"path":657,"priority":602},"references/example-research.md",{"path":659,"priority":602},"references/report-template.html",{"path":661,"priority":602},"references/research-patterns.md",{"path":663,"priority":602},"references/workflow.md",{"path":665,"priority":611},"scripts/compile_report.mjs",{"path":667,"priority":611},"scripts/extract_page.mjs",{"path":669,"priority":611},"scripts/list_urls.mjs",{"path":671,"priority":611},"scripts/package.json",{"basePath":147,"description":143,"displayName":145,"installMethods":673,"rationale":674,"selectedPaths":675,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/cookie-sync/SKILL.md",[676,677,678,679,680],{"path":595,"priority":561},{"path":599,"priority":597},{"path":601,"priority":602},{"path":604,"priority":602},{"path":681,"priority":611},"scripts/cookie-sync.mjs",{"basePath":169,"description":165,"displayName":167,"installMethods":683,"rationale":684,"selectedPaths":685,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/event-prospecting/SKILL.md",[686,687,689,690,691,692,693,695,696,698,699,700],{"path":595,"priority":561},{"path":688,"priority":602},"references/event-platforms.md",{"path":657,"priority":602},{"path":659,"priority":602},{"path":661,"priority":602},{"path":663,"priority":602},{"path":694,"priority":611},"scripts/__fixtures__/stripe-snapshot.json",{"path":665,"priority":611},{"path":697,"priority":611},"scripts/extract_event.mjs",{"path":667,"priority":611},{"path":671,"priority":611},{"path":701,"priority":611},"scripts/recon.mjs",{"basePath":189,"description":185,"displayName":187,"installMethods":703,"rationale":704,"selectedPaths":705,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/fetch/SKILL.md",[706,707,708,709],{"path":595,"priority":561},{"path":599,"priority":597},{"path":601,"priority":602},{"path":604,"priority":602},{"basePath":210,"description":207,"displayName":110,"installMethods":711,"rationale":712,"selectedPaths":713,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/functions/SKILL.md",[714,715,716],{"path":595,"priority":561},{"path":599,"priority":597},{"path":604,"priority":602},{"basePath":233,"description":229,"displayName":231,"installMethods":718,"rationale":719,"selectedPaths":720,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/safe-browser/SKILL.md",[721,722,723,725,727],{"path":595,"priority":561},{"path":599,"priority":597},{"path":724,"priority":611},"templates/claude-agent-sdk/.gitignore",{"path":726,"priority":611},"templates/claude-agent-sdk/hn-scraper-demo.mjs",{"path":728,"priority":611},"templates/claude-agent-sdk/package.json",{"basePath":252,"description":248,"displayName":250,"installMethods":730,"rationale":731,"selectedPaths":732,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/search/SKILL.md",[733,734,735,736],{"path":595,"priority":561},{"path":599,"priority":597},{"path":601,"priority":602},{"path":604,"priority":602},{"basePath":272,"description":268,"displayName":270,"installMethods":738,"rationale":739,"selectedPaths":740,"source":564,"sourceLanguage":22,"type":24},{"claudeCode":15},"SKILL.md frontmatter at skills/ui-test/SKILL.md",[741,742,743,744,745,747,749,751,753,755,757,758],{"path":595,"priority":561},{"path":563,"priority":597},{"path":599,"priority":597},{"path":601,"priority":602},{"path":746,"priority":602},"references/adversarial-patterns.md",{"path":748,"priority":602},"references/browser-recipes.md",{"path":750,"priority":602},"references/design-consistency.md",{"path":752,"priority":602},"references/design-system.example.md",{"path":754,"priority":602},"references/exploratory-testing.md",{"path":756,"priority":602},"references/parallel-testing.md",{"path":659,"priority":602},{"path":759,"priority":602},"references/ux-heuristics.md",{"basePath":517,"displayName":761,"installMethods":762,"rationale":763,"selectedPaths":764,"source":564,"sourceLanguage":22,"type":61},"browserbase-skills",{"npm":761},"cli ecosystem detected at /",[765,767],{"path":766,"priority":561},"package.json",{"path":563,"priority":561},{"sources":769},[770],"manual",{"npmPackage":761},{"closedIssues90d":11,"description":773,"forks":506,"homepage":774,"openIssues90d":508,"pushedAt":509,"readmeSize":504,"stars":510,"topics":775},"Browserbase's official collection of agent skills to access the web.","https://www.browserbase.com/SKILL.md",[],{"downloads":11},{"classifiedAt":778,"discoverAt":779,"extractAt":780,"githubAt":780,"npmAt":781,"updatedAt":778},1778683460101,1778683454817,1778683457134,1778683458753,[492,108,54,87,218],{"evaluatedAt":515,"extractAt":46,"updatedAt":515},[],[786,816,844,871],{"_creationTime":787,"_id":788,"community":789,"display":790,"identity":795,"providers":800,"relations":810,"tags":812,"workflow":813},1778685949178.7886,"k175j0a2ttdtwfrzvz3gae0z2186njwq",{"reviewCount":11},{"description":791,"installMethods":792,"name":793,"sourceUrl":794},"SDD WORK-PIPELINE Agent — Requirements analysis & development 6-agent full pipeline with DAG-based orchestration and sliding window context management",{"claudeCode":793},"uc-taskmanager","https://github.com/davepoon/buildwithclaude",{"basePath":796,"githubOwner":797,"githubRepo":798,"locale":22,"slug":799,"type":518},"plugins/agents-uc-taskmanager","davepoon","buildwithclaude","agents-uc-taskmanager",{"evaluate":801,"extract":806},{"promptVersionExtension":27,"promptVersionScoring":28,"score":29,"tags":802,"targetMarket":36,"tier":37},[108,803,804,805,492],"development","pipeline","sdd",{"commitSha":39,"license":807,"plugin":808},"GPL-3.0",{"mcpCount":11,"provider":537,"skillCount":809},3,{"repoId":811},"kd719kw54vhmcscq7ckdp59fg586mnt6",[492,108,803,804,805],{"evaluatedAt":814,"extractAt":815,"updatedAt":814},1778687422231,1778685949178,{"_creationTime":817,"_id":818,"community":819,"display":820,"identity":825,"providers":829,"relations":837,"tags":840,"workflow":841},1778675056600.2056,"k17ecb0pt8ntj8dbttr3x3frxs86mwdd",{"reviewCount":11},{"description":821,"installMethods":822,"name":823,"sourceUrl":824},"Create polished demo videos from screenshots and scene descriptions. Orchestrates playwright, ffmpeg, and edge-tts to produce product walkthroughs, feature showcases, and marketing teasers with story structure, scene design system, and narration guidance.",{"claudeCode":823},"demo-video","https://github.com/alirezarezvani/claude-skills",{"basePath":826,"githubOwner":827,"githubRepo":828,"locale":22,"slug":823,"type":518},"engineering/demo-video","alirezarezvani","claude-skills",{"evaluate":830,"extract":835},{"promptVersionExtension":27,"promptVersionScoring":28,"score":83,"tags":831,"targetMarket":36,"tier":37},[832,108,218,833,834],"video-creation","ffmpeg","tts",{"commitSha":39,"license":40,"plugin":836},{"mcpCount":11,"provider":537,"skillCount":535},{"parentExtensionId":838,"repoId":839},"k17dce6sbramb6sxm7ksr3928x86ncfs","kd7ff9s1w43mfyy1n7hf87816186m6px",[108,833,218,834,832],{"evaluatedAt":842,"extractAt":843,"updatedAt":842},1778675589331,1778675056600,{"_creationTime":845,"_id":846,"community":847,"display":848,"identity":853,"providers":858,"relations":865,"tags":867,"workflow":868},1778695660274.4905,"k1713acj4r8978e700nfa857wh86nszf",{"reviewCount":11},{"description":849,"installMethods":850,"name":851,"sourceUrl":852},"Relay PostHog events to Salesforce",{"claudeCode":851},"Salesforce Export Plugin","https://github.com/PostHog/posthog",{"basePath":854,"githubOwner":855,"githubRepo":856,"locale":22,"slug":857,"type":518},"nodejs/src/cdp/legacy-plugins/_destinations/salesforce","PostHog","posthog","salesforce",{"evaluate":859,"extract":864},{"promptVersionExtension":27,"promptVersionScoring":28,"score":860,"tags":861,"targetMarket":36,"tier":154},85,[856,857,862,863,87],"integration","data-export",{"commitSha":39},{"repoId":866},"kd7f22zf7qb3eschtk9s2qdv4586mfts",[87,863,862,856,857],{"evaluatedAt":869,"extractAt":870,"updatedAt":869},1778695919785,1778695660274,{"_creationTime":872,"_id":873,"community":874,"display":875,"identity":879,"providers":882,"relations":890,"tags":891,"workflow":892},1778695660274.4873,"k175zzq7h9kge7bgt0ntjtg3bx86npbf",{"reviewCount":11},{"description":876,"installMethods":877,"name":878,"sourceUrl":852},"Send event data and emails into Customer.io.",{"claudeCode":878},"Customer.io",{"basePath":880,"githubOwner":855,"githubRepo":856,"locale":22,"slug":881,"type":518},"nodejs/src/cdp/legacy-plugins/_destinations/customerio","customerio",{"evaluate":883,"extract":889},{"promptVersionExtension":27,"promptVersionScoring":28,"score":213,"tags":884,"targetMarket":36,"tier":154},[885,886,887,888,87],"customer-io","analytics","email","data-pipeline",{"commitSha":39},{"repoId":866},[886,87,885,888,887],{"evaluatedAt":893,"extractAt":870,"updatedAt":893},1778695677911]