Autotask — Claude Code MCP server: features and setup
Connect an AI agent to Autotask PSA to find tickets, log time, and work with projects through natural-language requests.
claude mcp add --env AUTOTASK_USERNAME=<AUTOTASK_USERNAME> --env AUTOTASK_SECRET=<AUTOTASK_SECRET> --env AUTOTASK_INTEGRATION_CODE=<AUTOTASK_INTEGRATION_CODE> --env AUTOTASK_API_URL=<AUTOTASK_API_URL> --env MCP_TRANSPORT=<MCP_TRANSPORT> --transport stdio autotask -- docker run -i --rm -e AUTOTASK_USERNAME -e AUTOTASK_SECRET -e AUTOTASK_INTEGRATION_CODE -e AUTOTASK_API_URL -e MCP_TRANSPORT ghcr.io/wyre-ai/autotask-mcp:v2.39.8What Autotask does
The server gives an agent tools for handling companies, contacts, tickets, projects, billing items, time entries, and notes. It supports searches and create or update operations for core records. In MCP Apps hosts, ticket details can also appear as an interactive card.
Setup uses an Autotask username, secret, and integration code, along with AUTOTASK_API_URL and MCP_TRANSPORT. Claude Code runs the server in Docker over stdio. The agent needs valid Autotask credentials to access the account.
What it helps with
- search tickets using filters
- record time against a ticket
- look up company and contact details
- update a project or ticket
Example prompts for the agent
- Find this company's open tickets and show their priorities.
- Add a time entry to this ticket.
- Show the ticket details with its contact and recent notes.
How to install Autotask
Platform CLI
claude mcp add --env AUTOTASK_USERNAME=<AUTOTASK_USERNAME> --env AUTOTASK_SECRET=<AUTOTASK_SECRET> --env AUTOTASK_INTEGRATION_CODE=<AUTOTASK_INTEGRATION_CODE> --env AUTOTASK_API_URL=<AUTOTASK_API_URL> --env MCP_TRANSPORT=<MCP_TRANSPORT> --transport stdio autotask -- docker run -i --rm -e AUTOTASK_USERNAME -e AUTOTASK_SECRET -e AUTOTASK_INTEGRATION_CODE -e AUTOTASK_API_URL -e MCP_TRANSPORT ghcr.io/wyre-ai/autotask-mcp:v2.39.8Replace <…> with your own values. 2 more environment variables are in the server README
Platform CLI
codex mcp add autotask --env AUTOTASK_USERNAME=<AUTOTASK_USERNAME> --env AUTOTASK_SECRET=<AUTOTASK_SECRET> --env AUTOTASK_INTEGRATION_CODE=<AUTOTASK_INTEGRATION_CODE> --env AUTOTASK_API_URL=<AUTOTASK_API_URL> --env MCP_TRANSPORT=<MCP_TRANSPORT> -- docker run -i --rm -e AUTOTASK_USERNAME -e AUTOTASK_SECRET -e AUTOTASK_INTEGRATION_CODE -e AUTOTASK_API_URL -e MCP_TRANSPORT ghcr.io/wyre-ai/autotask-mcp:v2.39.8Replace <…> with your own values. 2 more environment variables are in the server README
Config file
{
"mcpServers": {
"autotask": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"AUTOTASK_USERNAME",
"-e",
"AUTOTASK_SECRET",
"-e",
"AUTOTASK_INTEGRATION_CODE",
"-e",
"AUTOTASK_API_URL",
"-e",
"MCP_TRANSPORT",
"ghcr.io/wyre-ai/autotask-mcp:v2.39.8"
],
"env": {
"AUTOTASK_USERNAME": "<AUTOTASK_USERNAME>",
"AUTOTASK_SECRET": "<AUTOTASK_SECRET>",
"AUTOTASK_INTEGRATION_CODE": "<AUTOTASK_INTEGRATION_CODE>",
"AUTOTASK_API_URL": "<AUTOTASK_API_URL>",
"MCP_TRANSPORT": "<MCP_TRANSPORT>"
}
}
}
}Replace <…> with your own values. 2 more environment variables are in the server README
Config file
{
"servers": {
"autotask": {
"type": "stdio",
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"AUTOTASK_USERNAME",
"-e",
"AUTOTASK_SECRET",
"-e",
"AUTOTASK_INTEGRATION_CODE",
"-e",
"AUTOTASK_API_URL",
"-e",
"MCP_TRANSPORT",
"ghcr.io/wyre-ai/autotask-mcp:v2.39.8"
],
"env": {
"AUTOTASK_USERNAME": "<AUTOTASK_USERNAME>",
"AUTOTASK_SECRET": "<AUTOTASK_SECRET>",
"AUTOTASK_INTEGRATION_CODE": "<AUTOTASK_INTEGRATION_CODE>",
"AUTOTASK_API_URL": "<AUTOTASK_API_URL>",
"MCP_TRANSPORT": "<MCP_TRANSPORT>"
}
}
}
}Replace <…> with your own values. 2 more environment variables are in the server README
Platform CLI
gemini mcp add -e AUTOTASK_USERNAME=<AUTOTASK_USERNAME> -e AUTOTASK_SECRET=<AUTOTASK_SECRET> -e AUTOTASK_INTEGRATION_CODE=<AUTOTASK_INTEGRATION_CODE> -e AUTOTASK_API_URL=<AUTOTASK_API_URL> -e MCP_TRANSPORT=<MCP_TRANSPORT> autotask docker run -i --rm -e AUTOTASK_USERNAME -e AUTOTASK_SECRET -e AUTOTASK_INTEGRATION_CODE -e AUTOTASK_API_URL -e MCP_TRANSPORT ghcr.io/wyre-ai/autotask-mcp:v2.39.8Replace <…> with your own values. 2 more environment variables are in the server README
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about Autotask
How do I install Autotask in Claude Code?
Set the environment values and start the Docker server with `claude mcp add --env AUTOTASK_USERNAME=<AUTOTASK_USERNAME> --env AUTOTASK_SECRET=<AUTOTASK_SECRET> --env AUTOTASK_INTEGRATION_CODE=<AUTOTASK_INTEGRATION_CODE> --env AUTOTASK_API_URL=<AUTOTASK_API_URL> --env MCP_TRANSPORT=<MCP_TRANSPORT> --transport stdio autotask -- docker run -i --rm -e AUTOTASK_USERNAME -e AUTOTASK_SECRET -e AUTOTASK_INTEGRATION_CODE -e AUTOTASK_API_URL -e MCP_TRANSPORT ghcr.io/wyre-ai/autotask-mcp:v2.39.8`.
What does Autotask do?
It connects an AI agent to Autotask PSA, with tools to search and manage tickets, companies, projects, and other records.
What credentials does the Autotask connection need?
Provide an Autotask username, secret, and integration code. Configuration also uses AUTOTASK_API_URL and MCP_TRANSPORT.
