Weav MCP server
{
"mcpServers": {
"weav": {
"type": "http",
"url": "https://mcp.weav.com/mcp"
}
}
}{
"mcpServers": {
"weav": {
"type": "http",
"url": "https://mcp.weav.com/mcp"
}
}
}{
"mcpServers": {
"weav": {
"type": "http",
"url": "https://mcp.weav.com/mcp"
}
}
}{
"servers": {
"weav": {
"type": "http",
"url": "https://mcp.weav.com/mcp"
}
}
}{
"mcpServers": {
"weav": {
"type": "http",
"url": "https://mcp.weav.com/mcp"
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsonversion 1.0.1
What it does
The Weav MCP server connects AI tools to Weav customer service workspaces through OAuth authentication. It provides access to conversations, customers, and knowledge base management, allowing AI assistants to search support tickets, reply to customers, and manage training data within a workspace.
- Search and retrieve support conversations by subject, customer name, or email, then read recent message history to understand the context.
- Reply to customer conversations and update their status, priority, or assignee from within an AI tool.
- Search and manage knowledge base items (text, URLs, files, Q&A pairs) to keep customer-facing AI agents up to date with current information.
Tools this server exposes
| Tool | Does |
|---|---|
| search_conversations | Search support conversations by subject, customer name, or email with optional filters for status, channel, and spam. |
| get_conversation | Retrieve a conversation with its metadata and recent message history. |
| send_reply | Post a reply to a conversation as a workspace teammate. |
| update_conversation | Update a conversation's status, priority, or assignee. |
| search_customers | Search customers by name or email. |
| get_customer | Retrieve a customer profile, company information, and recent conversations. |
| search_training_data | Search knowledge base items by title, description, type, or source. |
| get_training_data | Retrieve a single knowledge base item. |
| add_training_data | Add knowledge base content as text, Q&A, or a website URL. |
| delete_training_data | Delete a knowledge base item. |
| resync_training_data | Re-scrape an existing URL-based knowledge base item. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.