Skip to content
Levr API Documentation
Download OpenAPI specOpenAPI

Hand off to an AI agent

Copy instructions to your AI agent

Copy this guide into your AI agent. It links to the current REST API and MCP documentation.

You are helping someone integrate their system with Levr. Use Levr's published documentation and machine-readable contracts as the source of truth. Do not invent endpoints, fields, or MCP tools.

  1. Start with the Levr API documentation to choose between a server-to-server REST integration and a user-authorized MCP connection.
  2. For REST, read the Markdown REST API guide, choose the user’s task, then fetch only the relevant resource guide and operation Markdown. Resource indexes include task guidance. For funding recommendations, start with the Matchmaking guide. Use the OpenAPI specification for exact paths, authentication, parameters, and envelope schemas.
  3. For REST client intake, Deal applications, or project Criteria, fetch that resource's contract_url. Its Application Contract is the authority for the current questions, stable keys, repeatable groups, file-transfer steps, accepted answer shapes, conditional visibility, supported operation URLs, and integration instructions. Repeatable fields require the returned item_id. File bytes use the expiring upload_url and multipart field named file; they never belong in JSON. Do not substitute a static example or the latest application definition.
  4. For an AI client acting with a signed-in user's access, read the MCP connection instructions in Markdown first, then the Model Context Protocol guide if needed. Let the connected MCP server expose its current tools and input contracts; do not infer tools from the REST API. Read the relevant MCP resource guide and follow the discovered tool’s revision, file-transfer, and confirmation requirements.
  5. Use URLs from the record handoff when supplied; otherwise use this documentation environment. Keep API keys and file-transfer credentials out of browser code, logs, prompts, and source control.
  6. Before implementing a write, verify authentication, required inputs, idempotency, error responses, and retry guidance in the current contract. If an operation is not documented, do not assume it exists.

Early access

Request an integration

Tell us what your integration needs to do. Add an email if you want a reply.

User-authorized AI clients

Model Context Protocol

Let compatible AI clients work with the same Levr access as the person who signs in.

Server URL
https://mcp.levr.ai/mcp
Authorization
OAuth with PKCE
Access
Your signed-in Levr account

Read the connection guide as Markdown.

Connect your AI client to Levr MCP

Connect to https://mcp.levr.ai/mcp using remote Streamable HTTP MCP. Sign in with your Levr account through OAuth with PKCE. An External API key, the REST API URL, and a bare /mcp path are not MCP connection credentials.

Before you start

Use the URL for the intended Levr environment. Hosted clients need a reachable HTTPS URL; localhost works only for a client on the same computer. Your client must support remote HTTP MCP and interactive OAuth. Your plan or workspace administrator may restrict custom connections.

Choose your client

Client instructions checked against the linked vendor documentation on 2026-10-01. Labels and availability can change; use the vendor link if your interface differs.

ChatGPT / ChatGPT Work

You can connect Levr from ChatGPT on the web; the desktop app is not required. Open the Plugins directory, which is separate from Settings → Plugins (https://chatgpt.com/settings/plugins-settings). The Settings page manages installed plugins and permissions; select Browse directory there to reach the directory.

  1. Open ChatGPT Plugins directly in your browser.
  2. Select the plus button, then Add custom MCP server. Name the connection Levr and enter https://mcp.levr.ai/mcp as the Server URL.
  3. Choose OAuth, sign in to Levr, and approve access. Use the default OAuth client setup; do not enter an External API key or invent a client secret.
  4. Review the risk warning and select Create as a plugin.
  5. Find Levr in your personal or workspace plugins, open it, and select the plus button to install it.
  6. Start a new chat, type @ in the prompt box, and select Levr. Use the read-only verification below to check the connection.

The current Plugins flow does not require a Developer mode step. Some accounts still show Settings → Apps → Advanced settings → Developer mode and Create app. If Add custom MCP server is missing from the Plugins directory, check your account's availability and workspace permissions, including Lockdown restrictions.

ChatGPT instructions checked on 2026-10-08. Sources: OpenAI custom MCP setup, plugin installation and use.

Codex

Run:

codex mcp add levr --url https://mcp.levr.ai/mcp
codex mcp login levr

Complete the browser sign-in. Use /mcp in Codex to inspect the active server. In the IDE extension, MCP servers → Add server also supports Streamable HTTP and an Authenticate action.

Source: Codex MCP setup.

Claude

Open Customize → Connectors → Add → Add custom connector, name it Levr, and enter https://mcp.levr.ai/mcp. Connect and complete Levr's browser authorization. Team and Enterprise workspaces may require an owner or authorized administrator to add the connector first.

Source: Claude custom connectors.

Claude Code

Run claude mcp add --transport http levr https://mcp.levr.ai/mcp, then use /mcp to authenticate with Levr.

Source: Claude Code MCP setup.

VS Code / GitHub Copilot

Run MCP: Add Server from the Command Palette, choose HTTP, and enter https://mcp.levr.ai/mcp. Follow the trust and OAuth sign-in prompts. These instructions cover interactive VS Code; other Copilot surfaces have separate capabilities.

Source: VS Code MCP servers.

Cursor

Add this entry to your personal ~/.cursor/mcp.json or the project's .cursor/mcp.json, then complete the client's OAuth sign-in:

{"mcpServers":{"levr":{"url":"https://mcp.levr.ai/mcp"}}}

Source: Cursor MCP integrations.

Gemini CLI

Add a levr entry under mcpServers in your Gemini CLI settings with "httpUrl": "https://mcp.levr.ai/mcp". Use /mcp auth levr to complete OAuth sign-in.

Source: Gemini CLI MCP servers.

Verify the connection

Ask your client: “List the Levr Portals I can access. Do not change anything.” A successful read through Levr confirms the connection works. A configured URL or successful authorization alone does not confirm that a tool call succeeded. An empty Portal list can be a valid result for an account with no Portal access.

Your client discovers available tools from Levr. Do not infer tool names from the REST API or configure a saved tool catalog. The connection uses your current Levr permissions.

Troubleshoot or disconnect

  • Missing custom-connection controls: check your client plan and workspace policy.
  • Sign-in or discovery failure: check the complete environment-specific URL, then reconnect and authorize again.
  • Missing tools or Portals: check the signed-in Levr account and its permissions.
  • Continuing failure: contact Levr support with the client name, URL, and error. Do not send passwords, authorization codes, tokens, or secrets.

Review or revoke access in Levr Account settings → Security → Connected MCP apps.

Give your AI a goal

Open Connect on a business, Project, or Deal and copy its instructions into your connected AI client. Add the outcome you want—for example, “Find the missing answers in this application and help me complete them. Do not submit it.” The instructions identify the record and link to its documentation.

Choose your task

Your live connection determines which tools are available. Read the guide and tool contracts relevant to your task instead of loading every contract.

Discover available tools

Section titled “Discover available tools”

Your MCP client discovers tools with tools/list and follows nextCursor until discovery is complete. To choose a Portal, use providers.list when available and pass the returned Portal slug as provider_slug where required. Follow collection cursors for further pages.

Read only the resource or tool reference relevant to your task. The live connection determines which tools your account can use; do not infer MCP tool names from REST endpoints. Before changing a record, read its current state and follow the tool’s revision and confirmation requirements. Reuse an idempotency key only for an identical retry.