Chatvolt

You can extend AI agents with a dedicated MCP server that exposes a curated set of tools to interact with the Chatvolt platform. This server acts as a bridge between your AI model and the Chatvolt API, enabling automated workflows, agent management, CRM scenarios, and datastore operations through simple tool calls.

How to use

Install and run the MCP server, then connect your application or agent runner to the provided MCP endpoint. You will discover the available tools, call them to perform actions, and receive structured results that you can incorporate into your automation flows.

How to install

# Prerequisites: Node.js (recommended LTS) and npm
node -v
npm -v

# 1) Install dependencies and build (if required by the project)
npm install
npm run build

# 2) Start or run the MCP server locally via the specified CLI
# Example: using npm script or a direct command as documented by the project
npm run start

Configuration and run notes

Configure your client to connect to the MCP server by specifying the command to launch the server and any required environment variables. If you are using the standard local setup, you will typically provide a command and environment values so that your client can start the server on demand.

Security and maintenance

Keep your API keys and credentials secure. Store secrets in a safe location and provide them to the MCP server only at runtime. Regularly rotate keys and monitor access to the MCP endpoint. Maintain dependencies with your usual security practices and apply updates as they are released.

Troubleshooting tips

If the MCP server fails to start, verify that your environment variables are correctly set, the required runtime (Node.js) is installed, and dependencies are installed. Check logs for error messages related to tool registration, service initialization, or network calls to the Chatvolt API.

Available tools

create_agent

Creates a new Chatvolt agent with name, description, modelName, systemPrompt, optional temperature, and an optional list of tools.

update_agent

Partially updates an existing agent by ID, allowing changes to one or more fields such as name, description, modelName, temperature, systemPrompt, and other configurations.

delete_agent

Deletes a specified agent by its unique ID.

list_agents

Retrieves a list of all available agents with no required parameters.

get_agent

Gets detailed information about a single agent by ID.

agent_query

Sends a query or message to an agent for processing, with optional conversation history.

enable_disable_agent_integration

Enables or disables a specific integration for an agent (e.g., WhatsApp, Telegram) by ID.

create_crm_scenario

Creates a new CRM scenario with a name and optional description.

update_crm_scenario

Updates an existing CRM scenario by ID, allowing changes to name and description.

delete_crm_scenario

Deletes a CRM scenario by ID.

list_crm_scenarios

Lists all CRM scenarios, optionally filtering by agentId.

create_crm_step

Creates a new step within a CRM scenario by providing scenarioId and step name.

update_crm_step

Updates an existing CRM step by step ID with a new name.

delete_crm_step

Deletes a CRM step by its ID.

list_crm_steps

Lists all steps for a given CRM scenario by scenarioId.

create_datastore

Creates a new datastore with a type, and optional name and description.

get_datastore

Retrieves information about a datastore by ID with an optional search term.

list_datastores

Lists all datastores available.

create_datasource

Creates a new data source within a datastore using datastoreId, a name, and the text content.