Browser Fu
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This instruction-only browser helper is coherent, but it encourages using your logged-in browser session and direct website API calls without tight scope or approval controls.
Install only if you are comfortable with the agent helping automate websites and inspect API traffic. Before use, require it to ask before using your logged-in browser profile, cookies, or authenticated/internal APIs, and keep the task limited to specific sites and read-only data unless you explicitly approve otherwise.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent could act as the logged-in user on websites and access private account data if invoked on a login-gated page.
This tells the agent to use the user's logged-in browser profile and session cookies. That is sensitive delegated account authority, but the metadata declares no credential requirement and the instruction is not limited to specific sites, accounts, or user-approved sessions.
Login wall | Try to automate login | Use `profile="user"` for existing session cookies
Require explicit user approval before using a logged-in browser profile or cookies, limit use to the named site and task, and avoid exposing cookies in prompts, logs, or command output.
The agent may make broad or authenticated requests to a website and retrieve more data than the user intended, potentially violating site rules or exposing private information.
The skill encourages raw API calls, cookie-authenticated requests, and bulk pagination. This is central to the skill's purpose, but it bypasses the safer UI workflow and does not clearly require read-only validation, throttling, or user approval for authenticated/internal APIs.
Once you find the endpoint, call it directly: ... curl -s "https://example.com/api/v1/items?page=1&limit=100" \ -H "Cookie: session=..." ... Loop through pages until you get all the data.
Ask the user before calling internal or authenticated APIs, confirm endpoints are read-only, throttle requests, avoid mutations, and document exactly what data will be collected.