# LLM Observability & Orchestration Agent

LLM Observability & Orchestration Agent is an MCP server that runs LangChain prompts on Gemini in Vertex AI, optionally traces them in LangSmith, and returns an execution result and trust manifest from self-test.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 2409 installs
- Author: a2a
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/a2a/llm-observability-orchestration
- Updated: 2026-05-31
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/llm-observability-orchestration-agent

## What it does

The `invoke` tool sends a prompt through a LangChain chain with system and human messages. It runs on Gemini in Vertex AI, and LangSmith tracing is optional. This gives an IDE agent a way to call the chain and inspect its execution.

Use `selftest` to run the agent against a fixed input. The response includes the result, an assertion grade and a trust manifest covering a signed Agent Card, JWKS, governance and licence details, and safety posture. Claims come with proof links. The server is accessed over MCP; configure its remote endpoint when connecting. `invoke` is tied to Gemini on Vertex AI.

## Requirements
- Remote server: `https://llm-observability-orchestration--a2a.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http llm-observability-orchestration https://llm-observability-orchestration--a2a.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add llm-observability-orchestration-agent --url https://llm-observability-orchestration--a2a.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "llm-observability-orchestration-agent": {
            "url": "https://llm-observability-orchestration--a2a.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "llm-observability-orchestration-agent": {
            "type": "http",
            "url": "https://llm-observability-orchestration--a2a.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http llm-observability-orchestration-agent https://llm-observability-orchestration--a2a.run.tools
```

## When to use it
- run a prompt through a LangChain chain
- send execution traces to LangSmith
- test the agent on a fixed input
- retrieve a trust manifest with proof links

## Example prompts
- Run my prompt through the LangChain chain on Gemini in Vertex AI using `invoke`.
- Use `selftest` and return the result with the trust manifest.
- Run the prompt with LangSmith tracing enabled.

## Similar
- [langsmith](https://affpapa.org/en/nevibecode/skills/langsmith.md) — skill · 328 installs
- [langsmith-experiment-diagnosis](https://affpapa.org/en/nevibecode/skills/langsmith-experiment-diagnosis.md) — skill · 184 installs
- [langchain-orchestration](https://affpapa.org/en/nevibecode/skills/langchain-orchestration.md) — skill · 453 installs
- [llm-app-development](https://affpapa.org/en/nevibecode/skills/llm-app-development.md) — skill · 139 installs
- [llm-application-dev](https://affpapa.org/en/nevibecode/plugins/seth-hobson-llm-application-dev.md) — plugin · ★ 126 repo stars
- [observability-llm-obs](https://affpapa.org/en/nevibecode/skills/observability-llm-obs.md) — skill · 4400 installs

---
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
