Vllora

v1.0.0Monitoring & Observabilitystable

Debug your AI agents

agentsai-agentsai-gatewayanthropicazure
Share:
805
Stars
0
Downloads
0
Weekly
0/5

What is Vllora?

Vllora is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to debug your ai agents

Debug your AI agents

This server falls under the Monitoring & Observability category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Debug your AI agents

Use Cases

AI agent debugging and monitoring
Multi-provider gateway support
vllora

Maintainer

LicenseNOASSERTION
Languagerust
Versionv1.0.0
UpdatedMay 22, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx vllora

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Vllora

Vllora is a lightweight real-time debugging and observability tool for AI agents that traces every LLM call, tool interaction, and agent workflow through an OpenAI-compatible API gateway running at localhost. It exposes a web UI on port 9091 and an API on port 9090, automatically collecting traces from frameworks like LangChain, Google ADK, and direct OpenAI SDK calls without requiring code changes. Developers use it to identify latency bottlenecks, inspect tool call sequences, and replay agent interactions during development.

Prerequisites

  • macOS with Homebrew installed (for the recommended install method)
  • API keys for the LLM providers you want to proxy (e.g. OpenAI, Anthropic)
  • An AI agent framework (LangChain, Google ADK, OpenAI SDK, etc.) to trace
  • An MCP client if using the MCP server integration feature
1

Install vllora via Homebrew

Add the vllora tap and install the binary.

brew tap vllora/vllora
brew install vllora
2

Start the vllora server

Launch the server — it starts the API on port 9090 and the debugging UI on port 9091.

vllora
3

Configure your LLM provider credentials

Open the web UI at http://localhost:9091 and add your API keys for each provider (OpenAI, Anthropic, etc.) through the settings interface.

4

Point your agent at the vllora endpoint

Change your agent's base URL from the provider's endpoint to vllora's local API. All calls will be proxied and traced.

# Example: redirect OpenAI calls through vllora
export OPENAI_BASE_URL=http://localhost:9090/v1
export VLLORA_OPENAI_API_KEY=your_openai_api_key
5

Inspect traces in the UI

Open http://localhost:9091 to see live traces, tool call sequences, latency breakdowns, and full request/response payloads for every agent interaction.

Vllora Examples

Client configuration

Claude Desktop config to run vllora as an MCP server for agent debugging.

{
  "mcpServers": {
    "vllora": {
      "command": "npx",
      "args": ["vllora"],
      "env": {
        "VLLORA_OPENAI_API_KEY": "your_openai_api_key"
      }
    }
  }
}

Prompts to try

Example ways to use vllora once connected via MCP.

- "Show me the last 10 traces captured by vllora"
- "Which tool calls had the highest latency in the last session?"
- "Replay the failed agent interaction from trace ID abc123"
- "List all LLM providers currently configured in vllora"

Troubleshooting Vllora

Port 9090 or 9091 is already in use

Find and stop the conflicting process with lsof -i :9090, or configure vllora to use different ports via its settings.

Agent requests are not appearing in the vllora UI

Verify that your agent's base URL is actually set to http://localhost:9090/v1 and not the original provider URL. Restart the agent process after changing the environment variable.

brew install vllora fails on Apple Silicon

Make sure Homebrew itself is up to date with brew update && brew upgrade, then retry. If the formula is not yet available for arm64, download the binary directly from the vllora GitHub releases page.

Frequently Asked Questions about Vllora

What is Vllora?

Vllora is a Model Context Protocol (MCP) server that debug your ai agents It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Vllora?

Follow the installation instructions on the Vllora GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Vllora?

Vllora works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Vllora free to use?

Yes, Vllora is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.

Browse More Monitoring & Observability MCP Servers

Explore all monitoring & observability servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "vllora": { "command": "npx", "args": ["-y", "vllora"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Vllora?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides