100Critical
This skill enables arbitrary browser control,
Behavior Analysis
Description verified — behavior matches claim
Claims to do
Agent Browser Skill: Fast browser automation using accessibility tree snapshots with refs for deterministic element selection.
Actually does
This skill executes the `agent-browser` command-line interface to perform headless browser automation. It navigates to specified URLs, takes accessibility tree snapshots in JSON format, interacts with web elements using generated references, and extracts data like text, HTML, or attributes. It also manages browser state (cookies, local storage), controls network requests, takes screenshots/PDFs, and supports isolated sessions and tab/frame management.
Skill Info
Namematrixy/agent-browser-clawdbot
Registryclawhub
Version0.1.0
PURLpkg:clawhub/matrixy/agent-browser-clawdbot@0.1.0
Stars318
Downloads87,490
Installs657
Source
SHA-2569643b9c65476...
SHA-1a3668f5c3af4...
MD5b13bea16d587...
TLSHfeb12fe5112c...
ssdeep96:yISlVdP+i...
Size5,548 bytes
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
ClawHubDocs
openclaw skills install matrixy/agent-browser-clawdbotAssessments (1)
AI Agent Traps ↗Broad Command Execution & Data Exfiltrationcriticalcommand execution
The skill provides extensive capabilities for executing arbitrary browser commands, including navigating to any URL, interacting with web elements, and critically, extracting various forms of data (text, HTML, attributes, cookies, local storage, session state). This poses a direct risk for data exfiltration, credential harvesting, and unauthorized web application manipulation.
100% confidenceline 106
agent-browser open <url>, agent-browser get text @e1, agent-browser cookies, agent-browser storage local key, agent-browser state save auth.json
Network Interception and Potential SSRFhighssrf
The `agent-browser network route` command allows for intercepting and modifying network requests, which can be used to bypass security measures or inject malicious content. The `agent-browser open <url>` command also enables navigation to arbitrary URLs, potentially facilitating Server-Side Request Forgery (SSRF) if the agent's environment allows access to internal networks.
90% confidenceline 120
agent-browser network route "**/api/*" --body '{"x":1}', agent-browser open https://example.comBadge
Markdown
[](https://mondoo.com/ai-agent-security/skills/clawhub/matrixy/agent-browser-clawdbot)HTML
<a href="https://mondoo.com/ai-agent-security/skills/clawhub/matrixy/agent-browser-clawdbot"><img src="https://mondoo.com/ai-agent-security/api/badge/clawhub/matrixy/agent-browser-clawdbot.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/clawhub/matrixy/agent-browser-clawdbot.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow