# Rius

Managed observability endpoint for AI agents. Lets coding assistants inspect traces, error groups, token costs, and alerts in chat.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI
- Author: glassflow
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/glassflow/rius-mcp-server
- Updated: 2026-10-22
- HTML: https://affpapa.org/en/nevibecode/mcp/rius

## What it does

Rius links your programming assistant directly to the GlassFlow telemetry platform over Streamable HTTP. It runs as a cloud service. You do not need to compile code or run local processes; connection goes to `https://mcp.eu.console.rius-glassflow.com/mcp`.

The toolkit includes methods to list spans, investigate failed executions, track spending by user, and update alert configurations. Interactive clients authenticate via OAuth 2.1 in your browser, while headless setups like CI pipelines accept an API token passed in the request header.

Connecting without an active workspace sending spans via the Rius SDK or OpenTelemetry yields nothing. Also, running with an API key blocks `ack_alert` and key generation outright, requiring full administrative browser login instead.

## Requirements
- Remote server: `https://mcp.eu.console.rius-glassflow.com/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http rius https://mcp.eu.console.rius-glassflow.com/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add rius --url https://mcp.eu.console.rius-glassflow.com/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "rius": {
            "url": "https://mcp.eu.console.rius-glassflow.com/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "rius": {
            "type": "http",
            "url": "https://mcp.eu.console.rius-glassflow.com/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http rius https://mcp.eu.console.rius-glassflow.com/mcp
```

## When to use it
- inspect failing agent runs directly inside the IDE
- compare latency and cost between release versions
- group recurring agent errors over the last day
- create notification rules when error thresholds spike

## Example prompts
- Find the agent with the highest error count in the past 24 hours.
- Show cost differences between version 1.4.0 and version 1.5.0 of support-agent.
- Set up an alert when checkout-agent failure rates exceed 5%.

## Similar
- [observability-driven-testing](https://affpapa.org/en/nevibecode/skills/observability-driven-testing.md) — skill · 1000 installs
- [genaispanmapper-mcp](https://affpapa.org/en/nevibecode/mcp/genaispanmapper-mcp.md) — MCP server
- [monitoring-specialist](https://affpapa.org/en/nevibecode/agents/monitoring-specialist.md) — subagent · 1677 installs
- [observability-engineer](https://affpapa.org/en/nevibecode/skills/observability-engineer.md) — skill · 93 installs
- [Superlog](https://affpapa.org/en/nevibecode/mcp/superlog.md) — MCP server · ★ 1477
- [application-performance-observability-engineer](https://affpapa.org/en/nevibecode/agents/application-performance-observability-engineer.md) — subagent · ★ 41887

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md