slbrow-skill

Browser automation via SLBrow MCP and HTTP API. Use when the user needs to navigate pages, analyze content, manage tabs, search history, extract text, or con...

This skill implements local browser automation and is internally consistent with its description. Before installing: 1) Verify you trust the SLBrow server and the browser extension it requires — they can read page content, selected text, and browser history. 2) Ensure the SLBrow server is bound to localhost and not exposed to remote networks. 3) Review the included script (scripts/slbrow.js) and tools.json if you want to confirm no external endpoints are contacted. 4) Be cautious using the skill with sensitive accounts or pages (banking, SSO flows) because the tools can extract page content and history. 5) If you prefer the agent not to call this automatically, disable autonomous invocation at the platform level or only enable this skill when needed.

Type: OpenClaw Skill Name: slbrow-skill Version: 1.0.0 The skill bundle provides browser automation tools with high-risk capabilities, specifically the `get_history` tool in `tools.json` which allows access to the user's full browser history. Furthermore, `SKILL.md` explicitly instructs the AI agent to use shell commands (`curl.exe`) to interact with a local REST API as a primary or fallback method, which increases the attack surface for shell injection or unauthorized local network interaction. While these features are aligned with the stated goal of browser control and Seelink video AI management, the inclusion of history access and the promotion of direct shell execution are significant security risks.