Agent-Browser on AWS AgentCore
vercel-labs/agent-browser
Runs agent-browser on cloud browser sessions hosted by AWS Bedrock AgentCore, with credentials taken from the environment or the AWS CLI.
Official agent skill
Agentcore Browser
by aws-samples in aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore
Browse web pages using a headless Chromium browser running inside the AgentCore container.
OfficialMIT-0Auto-check passed
Install Agentcore Browser
$ npx skills add aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore --skill agentcore-browser -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore agentcore-browser --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore.git skills-src && mkdir -p .claude/skills && cp -r skills-src/bridge/skills/agentcore-browser .claude/skills/agentcore-browser && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "agentcore-browser" agent skill from https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore/tree/main/bridge/skills/agentcore-browser into .claude/skills/agentcore-browser/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentcore-browser", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore/tree/main/bridge/skills/agentcore-browserType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore --skill agentcore-browser -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore agentcore-browser --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore.git skills-src && mkdir -p .agents/skills && cp -r skills-src/bridge/skills/agentcore-browser .agents/skills/agentcore-browser && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "agentcore-browser" agent skill from https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore/tree/main/bridge/skills/agentcore-browser into .agents/skills/agentcore-browser/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentcore-browser", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore --skill agentcore-browser -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore agentcore-browser --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/bridge/skills/agentcore-browser .cursor/skills/agentcore-browser && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "agentcore-browser" agent skill from https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore/tree/main/bridge/skills/agentcore-browser into .cursor/skills/agentcore-browser/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentcore-browser", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore.git --path bridge/skills/agentcore-browser--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore --skill agentcore-browser -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore agentcore-browser --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/bridge/skills/agentcore-browser .gemini/skills/agentcore-browser && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "agentcore-browser" agent skill from https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore/tree/main/bridge/skills/agentcore-browser into .gemini/skills/agentcore-browser/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentcore-browser", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore agentcore-browserInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore --skill agentcore-browser -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore.git skills-src && mkdir -p .github/skills && cp -r skills-src/bridge/skills/agentcore-browser .github/skills/agentcore-browser && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "agentcore-browser" agent skill from https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore/tree/main/bridge/skills/agentcore-browser into .github/skills/agentcore-browser/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentcore-browser", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore --skill agentcore-browser -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore agentcore-browser --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/bridge/skills/agentcore-browser .opencode/skills/agentcore-browser && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "agentcore-browser" agent skill from https://github.com/aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore/tree/main/bridge/skills/agentcore-browser into .opencode/skills/agentcore-browser/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentcore-browser", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
agentcore-browser- GitHub stars
- 177
- Token cost
- ~688 tokens
- SKILL.md length
- 241 words
- Files
- 5
- Skills in repo
- 5
- Repo updated
- First seen
- Licence
- MIT-0
At a glance
Browse web pages using a headless Chromium browser running inside the AgentCore container.
- The user asks to visit a website
- SKILL.md covers Important, Usage, From Agent Chat and Security Notes
- Runs JavaScript scripts from its folder; calls node
- Take a screenshot
What it does
Agentcore Browser is an agent skill from aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore, published by the product's own GitHub organization. Browse web pages using a headless Chromium browser running inside the AgentCore container. Navigate to URLs, take screenshots, and interact with page elements (click, type, wait, scroll). Requires enablebrowser=true in CDK config. Use when the user asks to visit a website, take a screenshot, fill in a form, or interact with a web page.
Its SKILL.md is about 690 tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files (for example `common.js`, `interact.js` and `navigate.js`).
The licence is MIT-0.
When your agent uses it
- The user asks to visit a website
- Take a screenshot
- Interact with a web page
Example prompts
- “/agentcore-browser”
Requirements
- Node.js
- Pre-approved tools (allowed-tools): Bash(node:*)
What it can do on your machine
Read from SKILL.md and the folder at commit b0c427f. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves these tools, so the agent can use them without asking each time:
Bash(node:*)
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships script files (JavaScript), which the agent can run.
Shell commands in SKILL.md call:
node
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Agentcore Browser loads about 688 tokens when it runs. Until then it costs about 89 tokens; SKILL.md has 241 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore at commit b0c427f, republished under its MIT-0 licence (© aws-samples). 241 words, ~688 tokens.
Download SKILL.mdSave it as
.claude/skills/agentcore-browser/SKILL.md (or your agent's skills folder). This skill also uses 4 other files; get the full folder from GitHub.- name
- agentcore-browser
- description
- Browse web pages using a headless Chromium browser running inside the AgentCore container. Navigate to URLs, take screenshots, and interact with page elements (click, type, wait, scroll). Requires enable_browser=true in CDK config. Use when the user asks to visit a website, take a screenshot, fill in a form, or interact with a web page.
- allowed-tools
- Bash(node:*)
AgentCore Browser
Headless Chromium browser running inside the AgentCore container. Navigate to URLs, take screenshots, and interact with page elements.
Important
This skill requires enable_browser=true in CDK configuration. If the browser is not available, the tools will return a clear error message.
Usage
browser_navigate
Navigate to a URL and return the page title and text content.
node {baseDir}/navigate.js '{"url": "https://example.com"}'url(required): The URL to navigate to
Returns JSON: {"url": "...", "title": "...", "content": "..."}
Content is truncated to 8000 characters to keep responses manageable.
browser_screenshot
Take a screenshot of the current browser page and send it to the user.
node {baseDir}/screenshot.js '{"description": "Homepage after login"}'description(optional): Caption for the screenshot
Returns text with [SCREENSHOT:{s3key}] marker that the proxy converts to an image.
browser_interact
Interact with the current page — click elements, type text, wait for elements, or scroll.
node {baseDir}/interact.js '{"action": "click", "selector": "#submit-btn"}'
node {baseDir}/interact.js '{"action": "type", "selector": "#search", "text": "hello"}'
node {baseDir}/interact.js '{"action": "wait", "selector": ".results"}'
node {baseDir}/interact.js '{"action": "scroll"}'action(required): One ofclick,type,wait,scrollselector(optional): CSS selector for the target element (required for click, type, wait)text(optional): Text to type (required for type action)
Returns JSON: {"success": true, "message": "..."}
From Agent Chat
- "Go to example.com" -> browser_navigate with url
- "Take a screenshot" -> browser_screenshot
- "Click the login button" -> browser_interact with action=click
- "Type my email in the form" -> browser_interact with action=type
- "Scroll down" -> browser_interact with action=scroll
- "Wait for the results to load" -> browser_interact with action=wait
Security Notes
- Browser runs inside an isolated per-user microVM
- Screenshots are uploaded to the user's S3 namespace (no cross-user access)
- The browser session file is stored at
/tmp/agentcore-browser-session.json - Navigation timeout: 30s, interaction timeout: 10s, wait timeout: 15s
© aws-samples, MIT-0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 4 other files in bridge/skills/agentcore-browser of aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore.
- SKILL.md
- common.js
- interact.js
- navigate.js
- screenshot.js
Open the folder on GitHubat commit b0c427f
Compare with similar skills
Agentcore Browser next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Agentcore Browser this skillaws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore | 177 | — | ~688 | Automated safety check: Pass | MIT-0 | |
| Agent-Browser on AWS AgentCorevercel-labs/agent-browser | 44k | — | ~1k | Automated safety check: Pass | Apache-2.0 | |
| Browseyc-software/qm | 15k | — | ~4k | Automated safety check: Pass | MIT | |
| Tiered Web Browsing and Scrapingcode-yeongyu/oh-my-openagent | 70k | — | ~2.7k | Automated safety check: Warn | Custom licence | |
| BrowseLeoYeAI/openclaw-master-skills | 2.2k | — | ~3.2k | Automated safety check: Notes | MIT | |
| Detecting Insider Threat Behaviorsmukul975/Anthropic-Cybersecurity-Skills | 34k | — | ~897 | Automated safety check: Pass | Apache-2.0 |
Similar skills
- Official
Browse
yc-software/qm
Drive a real stealth browser from your shell — act on websites (order food, file an expense, pull data behind a login), with per-person persistent sign-ins via the provider's managed auth (Kernel…
Tiered Web Browsing and Scraping
code-yeongyu/oh-my-openagent
Routes a web request through the cheapest tier that can finish it, from headless extraction with WAF bypass up to a real stealth or signed-in browser, with screenshots as proof.
Browse
LeoYeAI/openclaw-master-skills
Fast headless browser for QA testing and site dogfooding. An agent skill from LeoYeAI/openclaw-master-skills.
Detecting Insider Threat Behaviors
mukul975/Anthropic-Cybersecurity-Skills
Detect insider threat behavioral indicators including unusual data access, off-hours activity, mass file downloads, privilege abuse, and resignation-correlated data theft.
Performing Insider Threat Investigation
mukul975/Anthropic-Cybersecurity-Skills
Investigates insider threat incidents involving employees, contractors, or trusted partners who misuse authorized access to steal data, sabotage systems, or violate security policies, combining…
More from aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore
- Official
Clawhub Manage
aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore
Install, uninstall, and list ClawHub community skills. An agent skill from aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore.
- Official
Eventbridge Cron
aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore
Schedule recurring tasks using Amazon EventBridge Scheduler.
- Official
S3 User Files
aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore
Per-user persistent file storage backed by AWS S3. An agent skill from aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore.
- Official
API Keys
aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore
Dual-mode API key storage — native file-based or AWS Secrets Manager.
Questions about Agentcore Browser
What does Agentcore Browser do?
Browse web pages using a headless Chromium browser running inside the AgentCore container. Agentcore Browser is an agent skill from aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore, published by the product's own GitHub organization. Browse web pages using a headless Chromium browser running inside the AgentCore container.
When should I use Agentcore Browser?
Agentcore Browser fits situations like: the user asks to visit a website; take a screenshot; interact with a web page.
How do I install Agentcore Browser in Claude Code?
Run `npx skills add aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore --skill agentcore-browser -a claude-code`. Or copy the skill folder (bridge/skills/agentcore-browser in aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore) into .claude/skills/agentcore-browser in your project. Claude Code loads it when a task matches its description.
How do I install Agentcore Browser in Codex?
Run `npx skills add aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore --skill agentcore-browser -a codex`. Or copy the skill folder (bridge/skills/agentcore-browser in aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore) into .agents/skills/agentcore-browser in your project. Codex loads it when a task matches its description.
Can I use Agentcore Browser in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore --skill agentcore-browser -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/agentcore-browser, .gemini/skills/agentcore-browser, .github/skills/agentcore-browser and .opencode/skills/agentcore-browser in your project.
What does Agentcore Browser need to run?
Going by SKILL.md and its folder, Agentcore Browser needs JavaScript for the scripts in its folder and the command-line tools its instructions call (node). Our summary lists: Node.js. Its frontmatter pre-approves these tools: Bash(node:*).
Does Agentcore Browser access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Agentcore Browser safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Agentcore Browser use?
Agentcore Browser is published under the MIT-0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Agentcore Browser use?
About 688 tokens (SKILL.md is roughly 2.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Agentcore Browser?
Skills that share tags, products or a category with Agentcore Browser: Agent-Browser on AWS AgentCore (vercel-labs/agent-browser, 44k stars), Browse (yc-software/qm, 15k stars), Tiered Web Browsing and Scraping (code-yeongyu/oh-my-openagent, 70k stars) and Browse (LeoYeAI/openclaw-master-skills, 2.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Agentcore Browser?
aws-samples (a GitHub organization, an official publisher) maintains it in aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore, which has 177 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on September 28, 2026.
Source: aws-samples/sample-host-openclaw-on-amazon-bedrock-agentcore on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.