Agent Reach: One-Click Internet Access for Any AI Agent
Modern AI agents hit severe roadblocks when querying live social networks: commercial APIs charge hundreds of dollars per month, Reddit blocks cloud server IPs with 403 errors, and video/community platforms require login state. Agent Reach gives your AI Agent reliable, zero-API-cost internet access across 15+ platforms with automated backend health checking and failover routing.
❤️ Sponsors & Ecosystem Partners
Support open-source development · pnt01@foxmail.comExtracts any data from complex websites (Amazon, LinkedIn, X, Google Maps) in a real stealth browser with residential proxies and CAPTCHA handling.
Try Free (1,000 credits) ↗Deploy OpenClaw on Tencent Cloud Lighthouse in seconds, connect Agent Reach through chat, and add internet access to your agent setup.
Learn More ↗Web scraping platform with 100+ ready-made data collection tools for Amazon, TikTok, Google Maps, Instagram, and YouTube.
Free $3 Trial ↗One-click access to 200+ models, including leading open-source models such as Kimi K3, DeepSeek V3, and Qwen 2.5/3.
Explore Models ↗Why Do You Need Agent Reach?
AI Agents can already access the internet — but “can go online” is barely the start. The most valuable information lives across social and niche platforms where information density is highest, but each has strict barriers:
What You Might Want to Know Before Using
All tools are open source, all APIs used are free. The only possible cost is an optional server proxy ($1/mo) if hosting in restricted networks. Local machines need no proxy.
Cookies stay strictly on your local disk. Never uploaded to cloud servers. Fully open source code you can audit anytime.
Every platform routes through an ordered primary + fallback list. When an upstream path changes, Agent Reach auto-switches so you never notice.
Claude Code, OpenClaw, Cursor, Windsurf, Codex, Antigravity — any agent capable of running shell commands.
Supported Platforms (15)
Quick Start & Common Commands
Install Agent Reach: https://raw.githubusercontent.com/Panniantong/agent-reach/main/docs/install.mdagent-reach doctor
npx skills add Panniantong/Agent-Reach@agent-reach
agent-reach query reddit "best local LLMs"
agent-reach query twitter "Claude 3.7 benchmarks"
Design Philosophy & Core Principles
Agent Reach is a capability layer, not yet another tool. It sits one level above any specific implementation: it handles selection, installation, health checks, and routing, not the reading itself. Reading is done by your Agent calling upstream tools directly with zero wrapper overhead.
Each platform maintains an ordered list (primary + fallbacks). When an upstream tool is blocked (e.g. Bilibili 412 on yt-dlp), Agent Reach routes to bili-cli seamlessly.
agent-reach doctor tests live network responses, verifying that endpoints return actual content rather than merely checking if a CLI binary is installed.
Agent Reach never automates login or exfiltrates credentials. OpenCLI attaches only to an existing user-controlled desktop Chrome session.
Project Structure
channels/ ├── web.py → Jina Reader ├── twitter.py → twitter-cli ▸ OpenCLI ▸ bird ├── youtube.py → yt-dlp ├── github.py → gh CLI ├── bilibili.py → bili-cli ▸ OpenCLI ▸ search API (yt-dlp retired) ├── reddit.py → OpenCLI ▸ rdt-cli (login required) ├── facebook.py → OpenCLI (desktop browser session) ├── instagram.py → OpenCLI (desktop browser session) ├── xiaohongshu.py → OpenCLI ▸ xiaohongshu-mcp ▸ xhs-cli ├── linkedin.py → linkedin-mcp ▸ Jina Reader ├── rss.py → feedparser ├── exa_search.py → Exa via mcporter └── __init__.py → Channel registry (for doctor health checks)
Contributions to new platform channels, failover scrapers, and doctor diagnostics are welcome. Submit pull requests or report broken access paths on GitHub Issues.
Released under the MIT License. Free for both individual and commercial agentic pipelines.