| name | agent-reach |
| description | Give your AI agent one-click internet access — Twitter, Reddit, YouTube, GitHub, Bilibili, and more. Zero API fees. |
| version | 1.0.0 |
| author | Hermes Agent |
| license | MIT |
| platforms | ["linux","macos"] |
| metadata | {"hermes":{"tags":["agent","internet","web-access","research","twitter","reddit","youtube"],"related_skills":["crawl4ai","jina-reader"]}} |
Agent-Reach — Internet Access for AI Agents
One-click internet access layer for AI agents. Connects to Twitter/X, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu, LinkedIn, V2EX, and web search — all without API fees.
Prerequisites
- Python 3.10+
- pip
- Optional: proxy server for server-side usage ($1/month)
Installation
Agent-Reach is installed by telling your agent to fetch and run the install script:
Or manually:
pip install agent-reach
Usage
Health Check
agent-reach doctor
Shows what platforms work, what doesn't, and how to fix them.
Platform Setup
Each platform has its own setup command. Agent-Reach handles authentication, proxying, and fallback routing automatically.
In Agent Workflows
Once installed, your agent can:
- Read tweets and search Twitter/X
- Search Reddit and read threads
- Watch YouTube videos and get transcripts
- Browse GitHub repos, issues, and PRs
- Read Bilibili videos
- Search the web via Exa
Common Pitfalls
- Server-side usage may need a proxy — local computers don't
- Twitter/X requires a cookie for search functionality
- XiaoHongShu requires either OpenCLI (desktop) or MCP (server) setup
- Chinese documentation is more complete than English — check both
Verification
agent-reach doctor
Source