{
  "markdown": "# Loadster MCP\n\n[Loadster](https://loadster.com) is a cloud load testing and monitoring platform. Its built-in\n[Model Context Protocol](https://modelcontextprotocol.io/) (MCP) server lets AI agents like Claude, ChatGPT, Codex,\nCursor, and VS Code write and play load test scripts, build load test scenarios, set up monitors, and analyze\nresults in your Loadster account.\n\nThis repository is the public home for connecting to that server: setup snippets for each client, the Claude Code\nplugin, and the metadata behind Loadster's listings in MCP directories. The server itself is hosted by Loadster at\n`https://api.loadster.com/mcp`, and its source is not published here.\n\nThe maintained, full-length guide is the [AI Agents chapter of the Loadster manual](https://loadster.com/manual/ai-agents/).\nThis README is the short version.\n\n## The Loadster MCP endpoint\n\n| | |\n| --- | --- |\n| URL | `https://api.loadster.com/mcp` |\n| Transport | Streamable HTTP |\n| Authentication | OAuth 2.1 (preferred), or an MCP token in an `Authorization: Bearer` header |\n| Registry name | `com.loadster/loadster-mcp` |\n\n**OAuth** is the easiest way to connect. Clients that support MCP OAuth send you to Loadster to approve the\nconnection in your browser, and you can review or revoke connected agents on the **AI Agents** page in your Loadster\nsettings.\n\n**MCP tokens** are for clients that can't do OAuth, and for CI jobs and scripted agents. Create one in the Loadster\ndashboard under **Settings → AI Agents → MCP Tokens**, copy it right away (it's only shown once), and send it as\n`Authorization: Bearer YOUR_TOKEN`. A token acts as you within the team where you created it, so treat it like a\npassword.\n\n## Connecting Claude Code to Loadster\n\nThe quickest way is the plugin from this repository, which adds the Loadster MCP server plus a few skills that\nencode a sensible load testing workflow:\n\n```\n/plugin marketplace add loadster/loadster-mcp\n/plugin install loadster@loadster\n```\n\nThen run `/mcp`, pick **loadster**, and approve the OAuth connection in your browser. Ask Claude something like\n_\"List my Loadster projects\"_ to confirm it can reach your account.\n\nWithout the plugin, add the server directly and Claude Code will walk you through OAuth:\n\n```bash\nclaude mcp add --transport http loadster https://api.loadster.com/mcp\n```\n\nOr with an MCP token, for headless use:\n\n```bash\nclaude mcp add --transport http loadster https://api.loadster.com/mcp --header \"Authorization: Bearer YOUR_TOKEN\"\n```\n\nThe equivalent project-level `.mcp.json`:\n\n```json\n{\n  \"mcpServers\": {\n    \"loadster\": {\n      \"type\": \"http\",\n      \"url\": \"https://api.loadster.com/mcp\"\n    }\n  }\n}\n```\n\n## Connecting Claude Desktop and claude.ai to Loadster\n\nClaude Desktop and claude.ai connect to remote MCP servers as custom connectors, which use OAuth.\n\n1. Open **Settings → Connectors** and choose **Add custom connector**.\n2. Enter a name and the URL `https://api.loadster.com/mcp`.\n3. Choose **Connect** and approve the Loadster OAuth connection when your browser opens.\n4. In a conversation, open the tools menu and turn on the Loadster connector.\n\n## Connecting ChatGPT and Codex to Loadster\n\nChatGPT on the web connects through plugins created in **Developer mode** (**Settings → Security and login**).\nOpen **Plugins**, add a connection with the URL `https://api.loadster.com/mcp`, then add it to a new conversation\nfrom the tools menu and approve the OAuth connection.\n\nCodex in the ChatGPT desktop app, the Codex CLI, and the Codex IDE extension share one configuration:\n\n```bash\ncodex mcp add loadster --url https://api.loadster.com/mcp\ncodex mcp login loadster\n```\n\nOr with an MCP token in `~/.codex/config.toml`:\n\n```toml\n[mcp_servers.loadster]\nurl = \"https://api.loadster.com/mcp\"\nbearer_token_env_var = \"LOADSTER_MCP_TOKEN\"\n```\n\n## Connecting Cursor to Loadster\n\n[![Install in Cursor](https://cursor.com/deeplink/mcp-install-dark.svg)](cursor://anysphere.cursor-deeplink/mcp/install?name=loadster&config=eyJ1cmwiOiJodHRwczovL2FwaS5sb2Fkc3Rlci5jb20vbWNwIn0=)\n\nOr add it to `.cursor/mcp.json` in your project (or `~/.cursor/mcp.json` for all projects) and approve the OAuth\nconnection when Cursor first connects:\n\n```json\n{\n  \"mcpServers\": {\n    \"loadster\": {\n      \"url\": \"https://api.loadster.com/mcp\"\n    }\n  }\n}\n```\n\nTo use an MCP token instead, add `\"headers\": { \"Authorization\": \"Bearer ${env:LOADSTER_MCP_TOKEN}\" }` and set\n`LOADSTER_MCP_TOKEN` in the environment before starting Cursor.\n\n## Connecting VS Code to Loadster\n\n[![Install in VS Code](https://img.shields.io/badge/VS_Code-Install_Loadster_MCP-0098FF?style=flat-square&logo=visualstudiocode&logoColor=white)](https://vscode.dev/redirect/mcp/install?name=loadster&config=%7B%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fapi.loadster.com%2Fmcp%22%7D)\n[![Install in VS Code Insiders](https://img.shields.io/badge/VS_Code_Insiders-Install_Loadster_MCP-24bfa5?style=flat-square&logo=visualstudiocode&logoColor=white)](https://insiders.vscode.dev/redirect/mcp/install?name=loadster&config=%7B%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fapi.loadster.com%2Fmcp%22%7D)\n\nOr run **MCP: Add Server** from the Command Palette, or add it to `.vscode/mcp.json`:\n\n```json\n{\n  \"servers\": {\n    \"loadster\": {\n      \"type\": \"http\",\n      \"url\": \"https://api.loadster.com/mcp\"\n    }\n  }\n}\n```\n\nTo use an MCP token, declare a `promptString` input with `\"password\": true` and reference it in an\n`Authorization` header. The [manual](https://loadster.com/manual/ai-agents/#connecting-vs-code-to-loadster) has the\ncomplete example.\n\n## Connecting other MCP clients to Loadster\n\nAny client that supports remote servers over Streamable HTTP can connect the same way. For clients that only\nsupport local (stdio) servers, the [mcp-remote](https://www.npmjs.com/package/mcp-remote) bridge usually works.\nLeave out the `--header` arguments to have it run the OAuth flow instead of using a token:\n\n```json\n{\n  \"mcpServers\": {\n    \"loadster\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"mcp-remote\", \"https://api.loadster.com/mcp\", \"--header\", \"Authorization: Bearer YOUR_TOKEN\"]\n    }\n  }\n}\n```\n\n## What an agent can do with Loadster\n\nThe server exposes most of the Loadster dashboard. Clients fetch the live tool list when they connect, so this table\nis a map of the surface rather than the source of truth.\n\n| Group | Tools |\n| --- | --- |\n| Projects and documentation | `list_projects`, `get_documentation`, `get_scripting_api`, `get_example`, `list_command_types`, `get_command_schema`, `get_variable_schema` |\n| Scripts | `list_scripts`, `get_script`, `create_script`, `update_script`, `duplicate_script`, `delete_script`, `import_script`, `validate_script`, `play_script`, `get_play_status`, `stop_script`, `get_step_detail`, `get_screenshot`, `list_script_revisions`, `get_script_revision`, `restore_script_revision`, `list_script_assets`, `get_script_asset`, `put_script_asset`, `delete_script_asset` |\n| Datasets | `list_datasets`, `get_dataset`, `create_dataset`, `update_dataset`, `append_dataset_rows`, `delete_dataset` |\n| Scenarios and engines | `list_scenarios`, `get_scenario`, `create_scenario`, `update_scenario`, `delete_scenario`, `list_engines` |\n| Load test reports | `list_load_tests`, `get_load_test_report`, `update_load_test_notes` |\n| Monitoring | `list_monitors`, `get_monitor`, `create_monitor`, `update_monitor`, `disable_monitor`, `delete_monitor`, `list_monitor_cycles`, `get_monitor_cycle_detail`, `get_monitoring_summary`, `list_incidents`, `get_incident` |\n| Feedback | `submit_feedback` |\n\nSome actions are intentionally left to humans. An agent cannot launch or stop a full load test, enable a monitor,\nmanage notification policies or maintenance windows, or administer your team, billing, or Fuel. It gets everything\nready and hands off so you push the launch button.\n\nEvery tool carries MCP annotations (`readOnlyHint`, `destructiveHint`, `idempotentHint`, `openWorldHint`), so\nclients that honor them can prompt before writes and deletes.\n\n## Safety practices for AI load testing\n\n- **The agent acts in your account.** It edits the same scripts, scenarios, datasets, and monitors your team sees.\n  Script changes create revisions you can restore, but it's still a good idea to review proposed changes and keep\n  experiments in their own project.\n- **Only test what's yours.** Point scripts only at systems you own or are authorized to test. Playing a script runs\n  a single bot, but it's your responsibility either way.\n- **Scope your access.** Use one token per agent or machine, name them clearly, and revoke any you no longer use.\n  Review OAuth connections on the AI Agents page.\n\n## Support\n\n- Setup help and questions: [help@loadster.com](mailto:help@loadster.com)\n- Security issues: see [SECURITY.md](SECURITY.md)\n- Problems with the contents of this repository (the README, plugin, or directory metadata): open an issue here\n\n## What's in this repository\n\n| Path | Purpose |\n| --- | --- |\n| `server.json` | Loadster's entry in the [official MCP Registry](https://registry.modelcontextprotocol.io/) |\n| `glama.json` | Maintainer metadata for the [Glama](https://glama.ai/mcp/servers) listing |\n| `.claude-plugin/`, `.mcp.json`, `skills/` | The Claude Code plugin and its marketplace manifest |\n| `.github/workflows/` | Validation on every push, and registry publishing on release |\n\n## License\n\nThe contents of this repository are released under the [MIT License](LICENSE). The Loadster service itself is\ngoverned by [Loadster's terms of service](https://loadster.com/legal/terms-of-service/).\n",
  "bytes": 9579,
  "sha": "0ffc1754e44fd3950daefa0a7b66fedf490c10be7d6d1da11ca1186737b55f65",
  "repo_slug": "loadster/loadster-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_com_loadster_loadster_mcp_277923de/readme"
}