REPO PACKAGE
browser-act skills
COMMUNITY80skills
0ready to use
LABSCO SUMMARY
Two skills are the actual tool: browser-act, the CLI itself — stealth fingerprinting, CAPTCHA solving, account isolation, and a remote-handoff link a human can pick up when the agent gets stuck — and browser-act-skill-forge, which explores a site once and generates a new reusable skill from what it finds. The other 38 entries we can verify by name are what Skill Forge produces: nine skills for Amazon alone (search, ASIN lookup, reviews, buy-box monitoring, best-sellers, and two separate competitor-analysis skills), eight for Google (Maps search, Maps reviews, Maps contact extraction, image search, news, SERP, and social-profile discovery), five for Instagram, four each for generic e-commerce and Facebook, two for Goofish — China's largest second-hand marketplace — and one each for LinkedIn jobs, Indeed, Product Hunt, GitHub contributor lookup, business-contact and social-link discovery, and an 'industry key contact radar.'
That list only covers what sorts alphabetically ahead of 'producthunt-launches.' The README's own Solutions Catalog also promises Reddit, WeChat, YouTube, and Zhihu coverage among '30+ pre-built Skills', which fall past what we can verify here by name and description.
READ THE FULL ANALYSIS
Almost none of this works without a BrowserAct account. Only the two general tools are free to run with nothing else; every other skill in the package needs either an MCP server pointed at BrowserAct's service or an account key on file. The README's own pricing table shows that even the free tier — stealth browsing, CAPTCHA solving, remote handoff, Skill Forge itself — requires logging in, with only unlimited proxies and more than five concurrent stealth browsers actually gated behind payment.
This is the vendor's own repository, not a community collection, and its README reads like a product page rather than documentation. It pitches BrowserAct on four numbered 'things a browser for agents must get right' and offers 500 free credits to anyone who stars the repository and posts in its Discord — useful context for how to read the claims in it, since we have not independently tested BrowserAct's stealth or anti-bot claims ourselves.
WHAT'S INSIDE
80 showing · 80 total80 of 80
SETUP
Showing all 80 skills
amazon-alexa-qa
Types your questions into Amazon's own shopping assistant on the site and saves everything it answers, as text you can read through later.
amazon-asin-lookup-api-skill
Amazon gives every listing a code of its own, the ASIN; hand over one of those codes and the whole product record comes back — brand, price before and after any discount, star rating, description and listed features.
amazon-best-selling-products-finder-api-skill
Searches Amazon for a keyword and brings back the products that are selling best right now, not merely the ones that match the words.
amazon-buy-box-monitor-api-skill
When several sellers offer the same item on Amazon, one of them gets the Add to Cart button; this shows who holds it right now, at what price, and what the rival sellers are asking. The listing is named by its Amazon product code, the ASIN.
amazon-competitor-analyzer
Takes several rival Amazon products and works out how they compare on price, specs and what reviewers keep saying, as a written report plus a spreadsheet.
amazon-listing-competitor-analysis-skill
Pulls apart one competitor's Amazon page — the wording, the keywords, the photos, what reviewers complain about — and ends with where the opening is for your own product.
amazon-product-api-skill
Searches Amazon for a keyword, optionally just one brand, and collects each result's price, rating and physical details like color, material and style, page after page.
amazon-product-search-api-skill
Sweeps Amazon's search results for one keyword and brand, returning what each listing costs, how it is rated and roughly how many of them sell in a month.
amazon-reviews-api-skill
Fetches the customer reviews on one Amazon product — who wrote each one, how many stars, the full text, the date, and whether they were a verified buyer.
browser-act
Gives an AI assistant a real web browser it can drive — opening pages, clicking through them, filling in forms, taking screenshots and staying logged in between visits.
browser-act-skill-forge
The hard part of getting data out of a website is working out how the site hands it over; that happens once here, and the answer is kept as a tool you can run again any time.
business-contact-social-links-skill
Starts from a company name or its web address and tracks down where that business lives online — the official site, plus its LinkedIn, Facebook, X, Instagram, YouTube and TikTok pages.
ecommerce-listing
Reads a shopping site's search or category page and pulls out the products on it — name, price, picture, rating and link — carrying on through the following pages.
ecommerce-product-detail
Point it at one item in almost any online shop — by its link, a search word, or the barcode number on the box — and the whole listing comes off the page: price, brand, photos, stock, rating and the size or colour options.
ecommerce-reviews
Reads the review section of a product page on almost any online store and turns it into a list you can count and sort.
ecommerce-seller-info
Looks up who is behind a marketplace storefront — their name, how buyers rate them, how long they have been selling and what their returns policy is.
facebook-ads-library-search
Meta publishes every ad running on Facebook and Instagram, and this pulls out the ones from a given brand or a given search term — the wording, the pictures or video, the button, and the dates each ad ran.
facebook-groups-scrape-posts
Copies the posts out of a Facebook group into a list you can sort and count — who posted, what they wrote, the photos and videos, and how many reactions and comments each one drew.
facebook-page-posts
Turns a public Facebook Page's timeline into a table of posts — the text of each one, the links and hashtags inside it, and how many likes, comments and shares it collected.
facebook-page-profile-posts
A Facebook Page or a personal profile's posts, copied out with the images attached — each photo at full size, the poster's profile picture, and the names of anyone they posted alongside.
github-project-contributor-finder-api-skill
Finds the developers behind popular open-source projects — search GitHub by topic and star count, and get back each project's details together with its contributors' bios, personal sites and Twitter handles.
goofish-item-detail
Opens one listing on Goofish (闲鱼), China's big second-hand marketplace, and reads the whole page off it — the full write-up, the asking price, and every photo the seller posted.
goofish-search-list
Searches Goofish (闲鱼), China's biggest second-hand marketplace, page by page and writes down what is for sale: each item's title, its price, where the seller is, and how many people have said they want it.
google-image-api-skill
Runs a search on Google Images and saves the whole page of results — each picture, its caption, the site it came from, and the link to it — and it can search as though you were sitting in another country.
google-maps-api-skill
A contact list for local businesses, taken off Google Maps — search a trade in a given country and get each firm's name, address, phone number, website and star rating.
google-maps-contact-extract
Google Maps rarely shows a business's email address, so this carries on to the company's own website and picks it up there, together with the Facebook, Instagram or LinkedIn pages that site links to.
google-maps-reviews-api-skill
Collects what customers have written about a place on Google Maps — each review's star rating, its full text, who wrote it and when — so a few hundred can be read at once instead of scrolled through.
google-maps-search-api-skill
Sizes up a whole category of places on Google Maps: how pricey each one is, what sort of food or service it offers, what it has on site such as Wi-Fi, its star rating, and one of its reviews.
google-news-api-skill
A press-clippings run on Google News: name a subject and a stretch of time — the last few hours, the past week, the past year — and get back the headlines, who published them, when, and a link to each story.
google-search-serp
Takes down everything on one page of Google results for a search — the ordinary results in the order they appear, the ads above them, the related searches, and the AI summary when Google shows one.
google-social-media-finder
Type in a person's or a company's name and get back the social media accounts that come up under it — Instagram, X, TikTok, LinkedIn and the rest — each with its handle, its link and its follower count.
indeed-job-search
Collects job adverts off Indeed for a role and a place — the job title, the employer, the pay, the full description, the perks, and the link to apply.
industry-key-contact-radar-api-skill
Finds the people who hold a given job title in a given industry — founders in e-commerce, say — and brings back each name, the company behind it, a line about them, and a link to their profile.
instagram-hashtag-posts
Pick a hashtag and it gathers the posts Instagram is currently showing under it — the picture or video, the caption, the account behind it, and how many likes and comments each one drew.
instagram-place-posts
Finds an Instagram location by name and then collects the posts people have tagged there, either the ones Instagram ranks top or the most recent ones.
instagram-post-comments
Pulls the whole comment thread off one Instagram post — every comment, who wrote it, when, how many likes it got, and how many replies it drew.
instagram-profile-meta
The public facts on an Instagram account, read in one go: the bio, the follower and following numbers, how many posts there are, the link in the bio, and whether the account is verified or private.
instagram-profile-posts
Works through an Instagram account's grid a dozen posts at a time and saves each one — the caption, the picture or video, the date, where it was taken, and its like and comment counts.
linkedin-jobs-search
Sifts LinkedIn's job board down to the postings that fit — remote or on-site, full-time or contract, junior or senior, listed in the last day or the last month.
producthunt-launches
Product Hunt ranks the new products launched each day; this walks that chart for any day, week, month or year and brings back each product, the people who built it, and the email address on its website.
reddit-competitor-analysis-api-skill
Searches Reddit for a word or a brand name and brings back the discussions themselves — each post, the group it was posted in, how many upvotes it drew, and the replies underneath it, replies to replies included.
reddit-warmup
A 30-day routine that eases a new Reddit account into looking like an ordinary member — lurking and upvoting first, then commenting, then posting — before it starts working mentions of your product into the groups you pick.
social-media-finder-skill
Give it one name and it comes back with a spreadsheet of that person's or brand's accounts across Facebook, X, Instagram, LinkedIn and TikTok, each with its follower count and the first line of its bio.
taobao-keyword-search
Searches Taobao and Tmall for a product and writes down what each listing shows — the title, the price on the day, which shop is selling it, where it ships from, and how many people have bought it.
taobao-product-detail
Reads one Taobao or Tmall product page in full — what it costs now and what it cost before the discount, the shop selling it, every photo, the colour and size choices, and the specification table.
taobao-product-reviews
Collects the customer reviews off a Taobao or Tmall listing ten at a time — what each buyer wrote, when, which version of the item they bought, and the photos they attached.
taobao-shop-catalog
Walks through a Taobao or Tmall shop page by page and lists everything it has for sale, with the title and the photo of each item.
tiktok-hashtag-videos
Tells you how big a TikTok hashtag has become — how many videos carry it and how many views they have between them — then goes through the videos one page at a time, with the creator and the sound behind each.
tiktok-profile-videos
A TikTok account's whole posting history in one place — every video it has put out, with the views, likes and comments each one got and the sound behind it.
tiktok-search-videos
Search TikTok for a word and get the matching videos back as a list — who posted each one, how many views and likes it got, and the sound it uses.
tiktok-video-detail
Hand it the link to a single TikTok video and back comes everything that post carries — who made it, how it did, the sound, the hashtags, and any accounts tagged in it.
trustpilot-company-info
Looks up any company on Trustpilot from its web address alone — the rating, the number of reviews behind it, what kind of business it is, and how to contact it.
trustpilot-reviews
Reads the customer reviews on a company's Trustpilot page — the words people wrote, the stars they gave, where they are, and the company's reply. The list can be narrowed to a single star level, one language, or a recent date range.
web-research-assistant
Sends a research question out through BrowserAct's own browsers and brings back what it found with a source for each point — the fallback for when the sites you need will not open for an AI assistant.
web-search-scraper-api-skill
Any web page, turned into clean readable text from its address alone — the content itself, without the navigation and clutter around it.
webcrawler-deep-crawl
Walks a whole website from one starting address, page by page, and saves the readable text of each page it reaches, with the menus, headers and footers taken out.
wechat-article-search-api-skill
A lot of Chinese industry writing is published on WeChat's public accounts; search them by keyword and this returns each matching article in full — the whole body text, its author and the date it went out.
x-dm-auto-chat
Works your X (Twitter) inbox for you — finds the private messages still waiting on a reply, sends the answer your assistant wrote in the voice you set, and can also look up new people and message them first.
x-keyword-comment
Searches X (Twitter) for posts that mention a keyword and leaves a reply under each one, written in the voice you have set up for your brand.
x-tweet-by-conversation
Takes one post on X (Twitter) and collects the whole discussion underneath it — every reply, and the replies to those replies — with who wrote each one and how it did.
x-tweet-by-handle
Collects what a single X (Twitter) account has posted, straight from its profile — its own posts, its posts and replies together, or only the photos and videos it put up.
x-tweet-by-url
Hand it any X (Twitter) link — a search, a profile, a single post, or a list — and it works out what kind of page that is and pulls the posts off it.
x-tweet-search
Reads the tweets off whatever X (Twitter) page is already open in the browser — a search, a profile, a list — and turns what is on screen into a list you can save.
x-tweet-search-by-query
Runs a narrow search on X (Twitter) — a date range, a minimum number of likes, a language, a place — and keeps going through the results until the whole set is collected.
xiaohongshu-auto-posting
Runs the whole posting routine on a Xiaohongshu account: it looks at what is already getting attention in your niche, drafts a post in that style, publishes it once you approve, then checks back on how it did.
xiaohongshu-note-detail
Opens one Xiaohongshu post and reads off everything on it — the title and text, who wrote it, how many likes, saves and comments it has, its tags, and the comments themselves.
xiaohongshu-search
Type a keyword and get back the Xiaohongshu posts that match it — each one's title, author and like, save and comment counts, plus what is needed to open it in full afterwards.
xiaohongshu-user-profile
Everything a Xiaohongshu creator's page says about them — the bio, the follower count, the tags on the profile, the total likes and saves their posts have earned, and the list of posts itself.
youtube-api-skill
Searches YouTube for a keyword and pulls the numbers off every video that comes back — views, likes, comments, how long it runs, and the size of the channel behind it.
youtube-batch-transcript-extractor-api-skill
Every word spoken in a batch of YouTube videos, written out as text — the videos found by a keyword search, each one labelled with its title, its channel and its like count.
youtube-channel-api-skill
Finds the YouTube channels that come up for a search word, rather than their videos, and reports each one's name, handle, subscriber count, description and whether it carries a verified tick.
youtube-channel-business-email
Digs out the contact email a YouTube creator has published — first from the text on their About page, and failing that from the personal or business site the channel links to.
youtube-comments-api-skill
Searches YouTube for a term and collects what viewers wrote underneath each video that comes up — the comment itself, who left it, when, and how many likes and replies it drew.
youtube-influencer-finder-api-skill
A shortlist of YouTube creators working on a given topic, carrying both the numbers and the ways to reach them — subscribers, total views, country, a contact email when the channel shows one, and their other social links.
youtube-search-api-skill
Reads a YouTube results page for a search word and hands back what is listed on it — videos, Shorts, channels or playlists, each with its title, view count and date.
youtube-transcript
Turns what is said in a YouTube video into something readable — a summary, a chapter list, a social thread, a blog post, or just the standout quotes.
youtube-transcript-analysis-api-skill
Takes apart a rival's YouTube video without you having to watch it — what the opening does to hold people, who it is aimed at, the problems it claims to solve, how it sells, and what it leaves out.
youtube-transcript-extractor-api-skill
Give it the link to one YouTube video and it writes out everything said in it, along with the title, the channel that posted it and the like count.
youtube-video-api-skill
Lists what a single YouTube channel has published — newest first, most-watched first, or right back to its first upload — with each video's views, likes, comments and length.
zhihu-search-api-skill
Pulls whole articles off Zhihu for a keyword and a time window — the complete body text rather than a link and a snippet, plus the author, the date and the cover image.
No skill in this package matches that.
HOW TO GET IT
WHOLE PACKAGEInstall the whole package (80 skills)
npx skills add browser-act/skillsSINGLE SKILLOr install a single skill
npx skills add browser-act/skills --skill <name> --full-depthPick the skill name from the Skills tab — each entry there installs independently.