Skill collection · browserbase

browserbase/skills 18 graded agent skills from Browserbase

browserbase/skills brings together 18 skills for AI agent and LLM engineering, developer tools and frameworks, business, sales and finance, testing and QA, DevOps and cloud, productivity and knowledge work. Start with the task descriptions below, then compare the individual skill's score, source files and license.

View browserbase/skills on GitHub

At a glance

PublisherBrowserbase Official
Repositorygithub.com/browserbase/skills
Listed skills18
Security scanNot scanned
CategoriesAI Agents and LLM Engineering, Developer Tools and Frameworks, Business, Sales and Finance, Testing and QA, DevOps and Cloud, Productivity and Knowledge Work
LicenseMIT
Last commitSeptember 1, 2026
Install every skillnpx skills add https://github.com/browserbase/skills
README badgeGet the badge for this listing

What this collection offers

18 listed skills; recorded scores 70–84/100. 12 full reviews with separate installation details. Scores describe the saved source snapshot, using the Suede Skill Check.

Coverage: AI Agents and LLM Engineering (6) · Developer Tools and Frameworks (5) · Business, Sales and Finance (3) · Testing and QA (2) · DevOps and Cloud (1) · Productivity and Knowledge Work (1).

Snapshot: . The checks measure published files, documentation and maintenance signals. Security scan: not scanned; this is not a runtime test.

Choose a skill · Install and verify · All listed skills · Compare collections

Download the collection manifest: complete task descriptions, scores, license information, source revisions and install routes for all 18 listed skills. Use it to compare options or hand your selection to an agent.

Choose a skill for your task

Start with these 6 skills: the highest recorded score per category, then the highest remaining scores. The complete table below includes all 18 listings. Task descriptions come from the publisher; score explanations come from the saved report.

browser-use-to-stagehand 84/100 · B

Publisher's task description: Migrate browser-use (Python) browser-automation scripts to Stagehand v3 (TypeScript) on Browserbase. Use when the user wants to convert, port, rewrite, or migrate a browser-use Agent script to Stagehand, map browser-use features/APIs to Stagehand primitives (act/extract/observe/agent), or move agentic browser automation onto Browserbase with more determinism. Triggers on "browser-use", "browser_use", or "Agent(task=...)".

1,408 words in SKILL.md; 8 bundled files; includes reference material. Read the assessed SKILL.md.

Score evidence: Full points for structural checks, instruction depth and description length. Largest point gaps: curated-list citations (0/7); boundary clause (0/4).

See all score components for browser-use-to-stagehand
Structural checks56 of 56
Description length5 of 5
Boundary clause0 of 4
Example request3 of 3
Instruction depth7 of 7
Reference files3 of 3
Bundled scripts0 of 2
Recent maintenance6 of 9
Curated-list citations0 of 7
Tool permissions declared2 of 2
Compatibility declared2 of 2

License: MIT governs reuse and modification of these files.

  • Named environment variables: BROWSERBASE_API_KEY, ANTHROPIC_API_KEY.
  • Publisher compatibility declaration: The skill itself uses only Read/Write/Edit/Grep/Bash — no install step. The Stagehand code it generates needs Node 18+, `@browserbasehq/stagehand` (v3) and `zod`, plus `BROWSERBASE_API_KEY` / `BROWSERBASE_PROJECT_ID` and a model-provider key (e.g. `ANTHROPIC_API_KEY`) to run. The optional trace-assisted path uses the Browserbase SDK or the sibling `browser-trace` skill..
  • Requested tools: Read, Write, Edit, Grep, Bash.
Select one skill: browser-use-to-stagehand
npx skills add https://github.com/browserbase/skills/tree/main/skills/browser-use-to-stagehand

Download browser-use-to-stagehand score report · Read the full browser-use-to-stagehand review and install guide

company-research 84/100 · B

Publisher's task description: Company discovery and deep research skill. Researches a company's product and ICP, discovers target companies to sell to using Browserbase Search API, deeply researches each using a Plan→Research→Synthesize pattern, and scores ICP fit — compiled into a scored research report and CSV. Supports depth modes (quick/deep/deeper) for balancing scale vs intelligence. Use when the user wants to: (1) find companies to sell to, (2) research potential customers, (3) discover companies matching an ICP, (4) build a target company list, (5) do market research on prospects. Triggers: "find companies to sell to", "company research", "find prospects", "ICP research", "target companies", "who should we sell to", "market research", "lead research", "prospect list".

1,545 words in SKILL.md; 11 bundled files; includes reference material; includes a scripts folder. Read the assessed SKILL.md.

Score evidence: Full points for structural checks, instruction depth and example request. Largest point gaps: curated-list citations (0/7); boundary clause (0/4).

See all score components for company-research
Structural checks56 of 56
Description length3 of 5
Boundary clause0 of 4
Example request3 of 3
Instruction depth7 of 7
Reference files3 of 3
Bundled scripts2 of 2
Recent maintenance6 of 9
Curated-list citations0 of 7
Tool permissions declared2 of 2
Compatibility declared2 of 2

License: MIT governs reuse and modification of these files.

  • Named environment variables: BROWSERBASE_API_KEY.
  • Publisher compatibility declaration: Requires browse CLI (`npm install -g browse`) and BROWSERBASE_API_KEY env var.
  • Requested tools: Bash Agent.
Select one skill: company-research
npx skills add https://github.com/browserbase/skills/tree/main/skills/company-research

Download company-research score report · Read the full company-research review and install guide

add-webmcp 83/100 · B

Publisher's task description: Analyze an existing web application, identify safe user-visible capabilities across routes, forms, server actions, handlers, and schemas, then implement first-party WebMCP tools and validate discovery and invocation with Stagehand. Use when the user asks to make a codebase agent-ready, expose website features as WebMCP tools, or add WebMCP directly to an app rather than generating a standalone injection script from a URL.

1,611 words in SKILL.md; 22 bundled files; includes reference material; includes a scripts folder. Read the assessed SKILL.md.

Score evidence: Full points for structural checks, instruction depth and description length. Largest point gaps: curated-list citations (0/7); boundary clause (0/4).

See all score components for add-webmcp
Structural checks56 of 56
Description length5 of 5
Boundary clause0 of 4
Example request0 of 3
Instruction depth7 of 7
Reference files3 of 3
Bundled scripts2 of 2
Recent maintenance6 of 9
Curated-list citations0 of 7
Tool permissions declared2 of 2
Compatibility declared2 of 2

License: MIT governs reuse and modification of these files.

  • Named environment variables: BROWSERBASE_API_KEY.
  • Publisher compatibility declaration: Requires Node.js 22.18 or newer. Validation needs Chrome/Chromium locally or BROWSERBASE_API_KEY for a publicly reachable preview..
  • Requested tools: Bash Read Grep Edit Write.
Select one skill: add-webmcp
npx skills add https://github.com/browserbase/skills/tree/main/skills/add-webmcp

Download add-webmcp score report · Read the full add-webmcp review and install guide

browser-trace 80/100 · B

Publisher's task description: 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.

1,855 words in SKILL.md; 13 bundled files; includes a scripts folder. Read the assessed SKILL.md.

Score evidence: Full points for structural checks, instruction depth and description length. Largest point gaps: curated-list citations (0/7); boundary clause (0/4).

See all score components for browser-trace
Structural checks56 of 56
Description length5 of 5
Boundary clause0 of 4
Example request0 of 3
Instruction depth7 of 7
Reference files0 of 3
Bundled scripts2 of 2
Recent maintenance6 of 9
Curated-list citations0 of 7
Tool permissions declared2 of 2
Compatibility declared2 of 2

License: MIT governs reuse and modification of these files.

  • Named environment variables: BROWSERBASE_API_KEY.
  • Publisher compatibility declaration: Requires Node 18+, the browse CLI (`npm install -g browse`) with `browse cdp`, and optionally `jq` for ad-hoc querying of the bisected JSONL files. For remote Browserbase sessions, also requires `BROWSERBASE_API_KEY`. The skill scripts themselves use only the Node standard library — no `npm install` step..
  • Requested tools: Bash, Read, Grep.
Select one skill: browser-trace
npx skills add https://github.com/browserbase/skills/tree/main/skills/browser-trace

Download browser-trace score report · Read the full browser-trace review and install guide

functions 75/100 · C

Publisher's task description: 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.

297 words in SKILL.md; 3 bundled files. Read the assessed SKILL.md.

Score evidence: Full points for structural checks, description length and compatibility declared. Largest point gaps: curated-list citations (0/7); boundary clause (0/4).

See all score components for functions
Structural checks56 of 56
Description length5 of 5
Boundary clause0 of 4
Example request0 of 3
Instruction depth4 of 7
Reference files0 of 3
Bundled scripts0 of 2
Recent maintenance6 of 9
Curated-list citations0 of 7
Tool permissions declared2 of 2
Compatibility declared2 of 2

License: MIT governs reuse and modification of these files.

  • Publisher compatibility declaration: Requires the browse CLI (`npm install -g browse`), Node.js 18+ with pnpm or npm, and BROWSERBASE_API_KEY..
  • Requested tools: Bash Read Write Edit.
Select one skill: functions
npx skills add https://github.com/browserbase/skills/tree/main/skills/functions

Download functions score report · Browse the skill's source folder

Install and verify your selection

  1. Select one skill. Use the folder-specific command beside its description to start with the workflow you need. Skills with publisher-specific license terms link to that publisher's setup route.
  2. Read the setup requirements. Check the source, license and any named tools or environment variables. Installing a SKILL.md folder supplies instructions to the agent; any external CLI, MCP server, account or service still needs its own setup.
  3. Check the result. Open the installed SKILL.md and confirm that its name matches your selection. In Claude Code use the project's .claude/skills/ folder; in Codex use .agents/skills/. The skills CLI also supports other Agent Skills hosts.
  4. Try one representative task. Supply the input files, expected output and success criteria. Compare the result with those criteria before adding more skills or running the workflow on live work.

Step-by-step installation for Claude Code, Codex and Cursor · Read the directory's security and source policy.

Skills in browserbase/skills

Use the table to compare the full collection. Linked names open a detailed review when available; other names open the publisher's source folder.

SkillCategoryWhat it does
add-webmcp 83/100 · BAI agentsAnalyzes an existing web application for safe user-visible capabilities, implements first-party WebMCP tools, and validates discovery and invocation with Stagehand.

Get badges

Download report · Request a fresh review

agent-experience 74/100 · CAI agentsAudits the developer experience of a product, SDK, or docs site by sending Claude subagents at it with only a tiny task prompt.

Improve this score

Download report · Request a fresh review

autobrowse 83/100 · BAI agentsImproves browser automation through an iterative loop that runs a browsing task, reads the trace, and refines the navigation strategy until it reliably passes.

Get badges

Download report · Request a fresh review

browser-to-api 80/100 · BDev toolsTurns a website's observable HTTP traffic into a best-effort OpenAPI 3.1 spec by analyzing a browser-trace capture, to discover endpoints from a browser session.

Get badges

Download report · Request a fresh review

browser-trace 80/100 · BTestingCaptures a DevTools-protocol trace of browser automation, including CDP events, screenshots, and DOM dumps, and splits it into per-page searchable buckets for debugging.

Get badges

Download report · Request a fresh review

browser-use-to-stagehand 84/100 · BDev toolsMigrates browser-use Python automation scripts to Stagehand v3 TypeScript on Browserbase, mapping browser-use features and APIs to Stagehand primitives.

Get badges

Download report · Request a fresh review

company-research 84/100 · BBusinessResearches a company's product and ICP, discovers target companies with the Browserbase Search API, researches each deeply, and scores ICP fit.

Get badges

Download report · Request a fresh review

competitor-analysis 79/100 · CBusinessResearches competitors by auto-discovering them with the Browserbase Search API and studying each across a four-lane pattern, starting from the user's company.

Improve this score

Download report · Request a fresh review

event-prospecting 84/100 · BBusinessExtracts speakers from a conference page, filters their companies against the user's ICP, and researches ICP-fit speakers into a person-first HTML report.

Get badges

Download report · Request a fresh review

optimize-agent-prompt 81/100 · BAI agentsBuilds and improves Browserbase Agent API demos by running a fixed task, collecting logs, scoring results, revising one system-prompt heuristic, and confirming convergence.

Get badges

Download report · Request a fresh review

safe-browser 75/100 · CAI agentsBuilds constrained-browser agents whose safe_browser tool enforces a domain allowlist, so an agent can browse without raw browser, shell, or CDP access.

Improve this score

Download report · Request a fresh review

browser 78/100 · CDev toolsAutomate web browser interactions using natural language via CLI commands. Use when the user asks to browse websites, navigate web pages, extract…

Improve this score

Download report · Request a fresh review

fetch 78/100 · CDev toolsUse 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…

Improve this score

Download report · Request a fresh review

functions 75/100 · CDevOpsDeploy serverless browser automation as cloud functions using Browserbase. Use when the user wants to deploy browser automation to run on a schedule…

Improve this score

Download report · Request a fresh review

ui-test 70/100 · CTestingAI-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…

Improve this score

Download report · Request a fresh review

webmcp-gen 77/100 · CAI agentsCreate, compile, and validate site-specific WebMCP init scripts from a target URL and desired tool capability. Use when the user wants to author…

Improve this score

Download report · Request a fresh review

Compare before adding more skills

For AI agent and LLM engineering, compare task fit first, then the score components, installation requirements and license. Two skills with the same score can earn their points in different ways; use their reports to see the difference.

Compare AI agent and LLM engineering skills across publishers · Build a side-by-side comparison.

Show the listing in your README

This general badge links to the repository listing. For the two quality badges and feature offer, choose Get badges beside a skill below. Each skill keeps its own score and grade.

Suede AI Skills Directory: listed

Markdown
[![Suede AI Skills Directory: listed](https://suedeai.directory/browserbase/skills/badge.svg)](https://suedeai.directory/browserbase/skills/)
HTML
<a href="https://suedeai.directory/browserbase/skills/"><img src="https://suedeai.directory/browserbase/skills/badge.svg" alt="Suede AI Skills Directory: listed" height="20"></a>

Limited-time feature offer: Display either quality badge in your repository or skill README, linked to your listing, then request verification. Suede AI checks the rendered README at a recorded commit. Eligible skills scoring 80–99 with a verified badge can receive a featured placement while the offer is open. Request badge verification.

Where this repository is cited

browserbase/skills is listed as its publisher's official collection or from a listing request.