
by Loadster
Loadster MCP is the hosted MCP server of the Loadster load testing and monitoring platform. It lets AI assistants write and play scripts, build load test scenarios, set up monitors and read results.
PluginContact for pricing
Step 1: Choose how to authenticate: OAuth, or a personal access token created under Settings, API & MCP in the Loadster dashboard.
Step 2: In Claude Code, run claude mcp add --transport http loadster https://api.loadster.com/mcp, or install the Loadster plugin from the loadster-mcp repository.
Step 3: Run /mcp, pick loadster and approve the connection in your browser, choosing an access level.
Step 4: Ask the assistant to list your Loadster projects to confirm it can reach your account.
Step 5: Describe the flow to test, then have it write a script, build a scenario and read the results.
Step 6: Grant the Read, edit and execute level only if you want the assistant to launch tests itself.
The pages name Claude, ChatGPT, Codex, Cursor and VS Code. Clients need to support remote servers over Streamable HTTP and authenticate with OAuth or a Bearer header.
Only if you grant the Read, edit and execute access level. At the other two levels it prepares everything and you press the launch button.
The pricing page says the MCP server comes with every plan, including Loadster Fuel. Tests themselves burn fuel or count against your plan.
Revoke it on the API & MCP page in your Loadster settings. Changing your password also disconnects OAuth-connected assistants.
Requirements
2s.io MCP2s.io MCP exposes the 2s pay-per-call API catalog to AI agents as tools. Each call is paid in USDC through x402, with no account or API key.