# Agent Legs

> Agent Legs is the action execution layer for AI agents. Unified API for browsing, code execution, email, API calls, and file management — every action logged, permissioned, and auditable.

## Product

Agent Legs solves the action execution problem for AI agents. Most teams building AI agents spend significant time wiring together custom browser automation (Playwright/Selenium), code sandboxes (Docker containers), email senders (SMTP/transactional APIs), HTTP wrappers for third-party APIs, and file handlers. Agent Legs replaces all of this with a single SDK and hosted execution runtime.

Key capabilities:
- **Browse**: Full browser automation — click, scroll, extract, fill forms, handle auth
- **Run code**: Sandboxed Python, JavaScript, Bash execution with resource limits
- **Send email**: HTML/plain text email with attachments and threading
- **Call APIs**: Authenticated outbound HTTP with secret management and retries
- **Manage files**: Read/write/transform files across local and cloud storage

Every action is logged with inputs, outputs, and timestamps. Permission scopes prevent agents from taking actions outside their defined role.

## Pricing

- **Free**: 1,000 actions/month, all 5 action types, full audit log
- **Pro**: $69/month, 100,000 actions/month, audit log + replay, webhook notifications
- **Enterprise**: Custom volume pricing, dedicated infrastructure, SSO, custom SLA

## Target users

Developers and engineering teams building AI agents that need to take real-world actions. The product is most valuable when an agent needs to do more than answer questions — when it needs to browse a website, run a script, send a notification, call an external API, or read/write files.

## Pages

- [Home](https://agentlegs.com/)
- [Pricing](https://agentlegs.com/pricing.html)
- [Sign up](https://agentlegs.com/signup.html)
- [About](https://agentlegs.com/about.html)
- [Blog](https://agentlegs.com/blog/)
- [Privacy Policy](https://agentlegs.com/privacy.html)
- [Terms of Service](https://agentlegs.com/terms.html)

## Blog

- [The difference between an LLM and an AI agent](https://agentlegs.com/blog/llm-vs-ai-agent.html)
- [How to add browser automation to your AI agent without writing Selenium](https://agentlegs.com/blog/browser-automation-ai-agent.html)
- [Code execution in AI agents: the sandboxing problem nobody talks about](https://agentlegs.com/blog/code-execution-sandboxing.html)
- [Wiring AI agents to third-party APIs: a practical guide to tool calling](https://agentlegs.com/blog/wiring-agents-to-apis.html)
- [The action execution stack: what your AI agent needs to get work done](https://agentlegs.com/blog/action-execution-stack.html)