# AgentViva

> AgentViva is a free MCP server for developer tools. Uptime monitoring for AI tools, APIs and MCP servers, reachable from any MCP client. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Free
- **Platforms:** API, Web
- **Website:** https://agentviva.dev
- **Repository:** https://github.com/agentviva/docs
- **MCP Registry name:** dev.agentviva/agentviva
- **Categories:** developer tools, cloud and infrastructure
- **Listing updated:** 2026-11-14

## About

AgentViva answers one question per target: is it alive, and since when. It checks from the outside, the way a visitor or an AI app would, and keeps the evidence it saw, such as the page title, where redirects ended or the tools an MCP server listed. Four check types cover a web page, a JSON health endpoint with your own rules, an AI chat endpoint judged by a model, and an MCP server that must expose named tools.

The design leans against false alarms. A single failed probe changes nothing; the same result has to repeat roughly ten minutes apart before a state is recorded or a webhook fires, and the record says how many checks agreed. A target that refuses the checks is marked blocked rather than down, and no alert is sent. States are unknown, up, degraded, down and blocked.

Directories can connect a JSON or CSV catalogue of tools so each gets a monitor and a public status JSON. A tool that stays down for seven days, or a number you choose, lands in a decision queue where you can keep watching, call it a false alarm, stop watching or ask your site to remove it. AgentViva never changes your site itself; it sends your webhook a signed request. The service is invite-only early access, and its source is private.

## Features

- **Four check types:** Monitors pages, JSON health endpoints, AI chat answers and MCP servers.
- **Confirmed state changes:** Requires the same result to repeat before recording or alerting.
- **Proof with each state:** Stores the page title, redirect end point or MCP tool list that was seen.
- **Blocked is not down:** Targets that refuse checks are marked blocked and trigger no alert.
- **Catalogue sync:** Reads a daily JSON or CSV list so directories monitor many tools.
- **Decision queue:** Surfaces long-dead tools and lets you ask your site to remove them.

## Good for

- Watch a hosted MCP server and get alerted if a tool disappears
- Show a status badge on a README
- Check every listing in a directory for dead links
- Detect that a tool moved to a new domain
- Have an AI agent register monitors through MCP

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http agentviva https://agentviva.dev/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://agentviva.dev/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "agentviva": {
      "url": "https://agentviva.dev/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "agentviva": {
      "type": "http",
      "url": "https://agentviva.dev/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "agentviva": {
      "serverUrl": "https://agentviva.dev/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "agentviva": {
      "type": "http",
      "url": "https://agentviva.dev/mcp"
    }
  }
}
```

## Tools

- `list_decisions`: List monitored tools that need a decision (keep watching, false alarm, stop watching or remove)

## FAQ

### Is it free?

It is free during invite-only early access, with up to 25 monitors. Pro is coming soon with up to 10,000 monitors and AI canary checks; plans and prices will be published before anything is billed.

### Will a blocked site be reported as down?

No. A site that refuses the checks is marked blocked and no alert is sent.

### Can it remove listings from my site?

Only by request. It sends your webhook a signed remove_requested message and never changes your site itself.

### How long is history kept?

30 days today, with longer planned.

### Is AgentViva free?

Yes. AgentViva is free to use.

### How do I add AgentViva to Claude Code?

Run this command in your terminal: claude mcp add --transport http agentviva https://agentviva.dev/mcp

### What are the best alternatives to AgentViva?

The closest alternatives to AgentViva are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Ouroboros](https://ai-powered.com/mcp/ouroboros): Ouroboros is an open-source MCP server for developer tools.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Concord MCP](https://ai-powered.com/mcp/concord-mcp): Concord MCP is a freemium MCP server for developer tools.

---
Source: https://ai-powered.com/mcp/agentviva
