Web & Browser

A comprehensive browser automation testing skill for OpenClaw that performs functional, UI, accessibility, performance, and visual regression testing.

Email for AI agents. Create mailboxes, send and receive emails via API. No browser, no OAuth.

Deploy custom Juicebox hooks from browser. Compile Solidity, deploy contracts, verify on explorers, and attach to projects. Works with Claude-generated hooks.

Build RAG (Retrieval Augmented Generation) pipelines with web search and LLMs. Tools: Tavily Search, Exa Search, Exa Answer, Claude, GPT-4, Gemini via OpenRouter. Capabilities: research, fact-checking, grounded responses, knowledge retrieval. Use for: AI agents, research assistants, fact-checkers, knowledge bases. Triggers: rag, retrieval augmented generation, grounded ai, search and answer, research agent, fact checking, knowledge retrieval, ai research, search + llm, web grounded, perplexity alternative, ai with sources, citation, research pipeline

Parallel file download and optional tar extraction using the pget CLI (single URL or multifile manifest). Use when you need high‑throughput downloads from HTTP(S)/S3/GCS, want to split a large file into chunks for speed, or want to download and extract a .tar/.tar.gz in one step.

Use Perplexity API for web-grounded AI search and research. Use when user needs up-to-date information, multi-step reasoning with web citations, exhaustive research with source references, factual queries with current events, or competitive analysis. Default when user mentions Perplexity, needs current information, or requires source citations.

Temporary real-time chat rooms for AI agents. Password-protected, with SSE streaming, web UI for humans, and CLI tools for agents.

Publish viral X (Twitter) Articles with AI. Long-form content that gets engagement. Proven hook patterns, browser automation. Works with Claude, Cursor, OpenClaw.

Research any topic across Reddit, X/Twitter, and the web from the last 30 days. Synthesizes findings into actionable insights or copy-paste prompts.

Controls a local browser from a sandboxed remote machine. Use when the agent is running in a sandbox (no GUI) and needs to navigate websites, interact with web pages, fill forms, take screenshots, or expose local dev servers via tunnels.

Control the Linux desktop GUI using xdotool, wmctrl, and dogtail. Use when you need to interact with non-browser applications, simulate mouse/keyboard input, manage windows, or inspect the UI hierarchy of applications on X11/GNOME. Supports: (1) Clicking/typing in apps, (2) Resizing/moving windows, (3) Extracting text-based UI trees from apps (A11y), (4) Taking screenshots for visual analysis.

Optimize prompts for clarity and effectiveness. Use when user says "improve this prompt", "optimize my prompt", "make this clearer", or provides vague/unstructured prompts. Intelligently routes to sub-agents for codebase research, clarifying questions, or web search as needed.