Hermes

Hermes Agent is a self-improving agent from Nous Research with a built-in MCP client. Route its model calls through the AI Gateway, then connect it to the Orq MCP server for agentic access to the workspace.

AI Gateway

Route Hermes’s model calls through the AI Gateway for unified tracing and cost tracking.

MCP

Manage Agents, experiments, and Traces from natural language.

Prerequisites

Hermes installed (desktop app or CLI)
Active Orq.ai account
An Orq.ai API key

AI Gateway

Hermes is OpenAI-compatible, so it routes through the AI Gateway by pointing the OpenAI provider at the router. Configure it in the desktop app or the CLI. Changes made in the desktop app apply to the CLI as well, since both share the same configuration.

Setup

Desktop app
CLI

Open provider settings

Open the Hermes desktop app. On first launch, skip provider selection to set it later. Open Settings in the top right, then go to Providers > API keys.

Set the OpenAI API key and base URL

Scroll to OpenAI API and expand it. Paste the Orq.ai API key into the OpenAI API key field and click Save. Then paste the router URL into the OpenAI API base URL override field and click Save:

https://my.orq.ai/v3/router

Hermes API keys settings with the OpenAI API entry expanded, the Orq.ai key in the OpenAI API key field, and the router URL in the base URL override field — Overriding the OpenAI provider base URL with the Orq.ai router in Hermes

Select a model

Hermes now routes through the AI Gateway. Pick a model from the recommended list, or search by name for any other model from the AI Gateway catalog that is not shown.

Open the config with hermes config edit and set the model block to a custom provider pointed at the router, with the Orq.ai key inline:

model:
  provider: custom
  default: "anthropic/claude-sonnet-5"
  base_url: "https://my.orq.ai/v3/router"
  api_key: "<your-orq-api-key>"

Set default to any provider-prefixed model from the AI Gateway catalog. Override the model for a single session with hermes chat --model openai/gpt-5.4.

The default config already contains a model block. Edit its existing keys; the block must end up with a single provider and a single base_url.

Verification

Send a prompt in Hermes. The response appears in Hermes and the request appears in Orq.ai Traces with the selected model identifier.

MCP

The Orq MCP server exposes the Orq.ai platform as MCP tools. Hermes’s built-in MCP client connects to it and uses those tools to manage Agents, run experiments, and query Traces from natural language.

Installation

Desktop app
CLI

Open MCP settings

In Settings, go to MCP and click New server in the top right.

Add the Orq MCP server

Set the Name to orq and paste the following into the Server JSON field, then click Save server:

{
  "url": "https://my.orq.ai/v2/mcp",
  "headers": {
    "Authorization": "Bearer YOUR_ORQ_API_KEY"
  }
}

Replace YOUR_ORQ_API_KEY with an Orq.ai API key.

Hermes MCP settings showing a new server named orq with the Orq MCP URL and a bearer Authorization header in the Server JSON field — Adding the Orq MCP server in the Hermes desktop app

Reload MCP

Click Reload MCP in the top right. Hermes connects to the Orq MCP server and its tools become available.

Add the Orq MCP server to ~/.hermes/config.yaml under mcp_servers:

mcp_servers:
  orq:
    url: "https://my.orq.ai/v2/mcp"
    headers:
      Authorization: "Bearer <your-orq-api-key>"

Reload MCP servers without restarting by running /reload-mcp inside Hermes, then confirm the connection:

hermes mcp test orq

Available Commands

Once connected, ask Hermes to perform operations across the workspace:

Agents

Create an agent with custom instructions and tools
Get agent configuration for [agent-key]
Update agent [agent-key] with new instructions or model
Configure agent with evaluators and guardrails
Invoke agent [agent-key] with input [message]
Retrieve agent response [response-id]

Deployments

Create a deployment called [deployment-key]
Get deployment configuration for [deployment-key]

Skills

Create a skill called [skill-key]
List all skills in my workspace
Get skill [skill-key]
Update skill [skill-key]
Delete skill [skill-key]

Analytics

Get analytics overview for my workspace
Show me workspace metrics for the last 7 days
Query analytics filtered by deployment ID

Datasets

Create a dataset called "customer-queries"
List all datapoints in dataset [dataset-key]
Add datapoints to dataset [dataset-key]
Update datapoint [datapoint-id]
Delete specific datapoints in dataset [dataset-key]
Delete dataset [dataset-key]

Experiments

Create an experiment from dataset [dataset-key]
List all experiment runs
Export experiment run [run-id] as CSV
Run experiment and auto-evaluate results

Evaluators

Get evaluator configuration for [evaluator-key]
Create an LLM-as-a-Judge evaluator for tone
Create a Python evaluator to check response length
Add evaluator to experiment [experiment-key]
Update evaluator [evaluator-key] with a new prompt
Update Python evaluator [evaluator-key] with revised code

Traces

List traces from the last 24 hours
Show me traces with errors
Get span details for trace [trace-id]
Find the slowest traces from today
Show all traces for thread [thread-id]

Models

List all available chat models
List all available embedding models
Invoke model [model-id] with prompt [message]

Search for datasets named "customer"
Find experiments in project [project-id]
List directories in project [project-id]

Documentation

Search the Orq.ai docs for [topic]

Managing Entities

Delete agent [agent-key]
Delete experiment [experiment-key]
Delete evaluator [evaluator-key]
Delete prompt [prompt-key]
Delete knowledge base [knowledge-base-key]

Use delete_dataset to delete a dataset along with all its datapoints.

See the MCP Quickstart for the full tool reference and examples.

Skills

Orq Skills add pre-built agentic workflows to Hermes for the full Build, Evaluate, Optimize lifecycle.

Installation

npx skills add orq-ai/assistant-plugins --agent hermes-agent

Set an API key and connect the Orq MCP server first. Skills are then triggered by describing what is needed.

Orq Skills

The full catalogue of skills and slash commands.

Slash commands (/orq:quickstart, /orq:traces, and others) are only available in Claude Code.

Orq MCP Quickstart

The full Orq MCP tool reference.

API Keys

Create and manage Orq.ai API keys.

Model Providers

Coding Agents

Chat Interfaces

Frameworks & SDKs

Automations

AI Gateway

MCP

Prerequisites

AI Gateway

Setup

Verification

MCP

Installation

Available Commands

Skills

Installation

Orq Skills

See Also

Orq MCP Quickstart

API Keys

AI Gateway

MCP

​Prerequisites

​AI Gateway

​Setup

​Verification

​MCP

​Installation

​Available Commands

​Skills

​Installation

Orq Skills

​See Also

Orq MCP Quickstart

API Keys

Prerequisites

AI Gateway

Setup

Verification

MCP

Installation

Available Commands

Skills

Installation

See Also