rad quote MCP
Enables creating projects from uploaded files directly from AI assistants. Provides tools for obtaining upload URLs and creating project import jobs from previously uploaded files.
README
rad quote MCP
The rad quote MCP server lets you create projects from uploaded files directly from your AI assistant.
Pick the install path that matches you:
| Your situation | Path |
|---|---|
| I use Claude Desktop and do not want to install Node or open a terminal | Claude Desktop UI — paste https://api.limits.run/mcp?key=<YOUR_API_KEY> into the Custom Connector form |
| I am a developer, I have Node 18+, I use Cursor / Codex / Windsurf / Claude Code | One-command CLI |
| I want a double-click install for Claude Desktop and have Node 18+ on my machine | .mcpb bundle |
One-command CLI
npx @radquote/mcp-setup@alpha install --env=stage
Requires Node 18+. Detects every supported client on your machine and writes the configuration for it. By default it configures all of them; pass --client=<id> to target one, or --pick to choose interactively.
Supported clients: claude-code, claude-desktop, cursor, windsurf, codex.
More detail: packages/setup/README.md.
Manual config snippets
If you prefer to edit configuration files yourself, copy the snippet for your client from examples/. Replace <MCP_URL> with https://api.limits.run/mcp (stage) and <YOUR_API_KEY> with your rad quote workspace API key.
What you get
Three tools appear in your AI assistant:
get_upload_url— obtain a short-lived signed URL for uploading one file.create_project_import_job— create a new project by importing tasks from previously uploaded files; optionally copy rates from an existing project as a read-only template.get_project_import_job— look up the current status of an import job by id. When the job is complete, the response containsproject_idof the newly created project; when it has failed, the response carries a human-readableerror.
And one orchestration prompt:
estimate_project_workflow— a single entry point that walks the agent through the full flow: upload the file, create the import job, poll the job every 15 seconds, and reply with a link to the resulting project (or with the failure reason). Arguments:file(required absolute local path),title,client_title,minimum_budget,source_project_id(all optional).
Workflow: create a project from a file
The estimate_project_workflow prompt packages the full recipe so you do not have to spell out the steps to the agent. Invocation depends on the client:
- Claude Code:
/mcp__radquote-stage__estimate_project_workflow, then fill infileand the optional fields. - Claude Desktop / Cursor / Windsurf: pick
estimate_project_workflowfrom your client's prompts menu and fill the arguments. - Codex: invoke via your normal MCP prompt command.
The workflow requires the agent to PUT the file to a signed URL. That means it works in clients that can run shell commands (Claude Code, Cursor, Codex, Windsurf) but not in Claude Desktop's built-in chat, which cannot perform arbitrary HTTP uploads. In Claude Desktop, upload the file some other way and then invoke
create_project_import_jobwith its public URL directly.
For reference, the PUT step looks like:
curl -X PUT --upload-file ./my-file.xlsx \
-H "Content-Type: application/vnd.openxmlformats-officedocument.spreadsheetml.sheet" \
"<upload_url>"
The response is HTTP 201 with a JSON body containing url — feed that into create_project_import_job.file.
Status
Pre-release. The production endpoint URL and the MCP Registry listing are being finalized. Currently only the stage environment (https://api.limits.run/mcp) is wired up — intended for internal QA and partner integration. Running the CLI with --env=prod prints a clear "not configured" message until production ships.
See docs/stage.md for details on the stage environment.
License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
E2B
Using MCP to run code via e2b.