Quip Artifact Manipulation Backend

This Model Context Protocol (MCP) service provides an interface layer for generative AI entities (such as Claude) to programmatically engage with Quip documents.

Core Capabilities

• Retrieval: Securely fetch and return the contents of a specified Quip thread by its unique identifier.
• Content Ingestion (Suffix): Append novel textual material to the conclusion of an existing document body.
• Content Ingestion (Prefix): Prepend new textual material to the commencement of an existing document body.
• Content Substitution: Completely replace the extant content of a document with new data.
• Artifact Genesis: Planned functionality to provision new Quip documents (currently redirects interaction to the native web UI).

Operational Flow

The service acts as a secure conduit between the consuming AI framework and the Quip API. The process involves:

1. Intercepting instructions adhering to the MCP specification.
2. Invoking a backend Python utility (quip_edit_fixed.py) with correctly mapped parameters.
3. Relaying the operation outcome back to the requesting AI system.

Prerequisites for Deployment

• Runtime Environment: Node.js version 18 or greater.
• Development Framework: TypeScript setup.
• Python Dependencies: Python interpreter installed, including the necessary quip SDK package.
• Authentication Credential: A valid, authorized Quip personal access token.

Deployment Steps

1. Acquire the repository source code:

   git clone https://github.com/AvinashBole/quip-mcp-server.git cd quip-mcp-server

2. Install required node modules:

   npm install

3. Compile the source code:

   npm run build

4. Configure the MCP host environment variables/settings (typically within the client configuration):

   { "mcpServers": { "quip": { "command": "node", "args": ["path/to/quip-server/build/index.js"], "env": { "QUIP_ACCESS_TOKEN": "your-quip-access-token", "QUIP_BASE_URL": "https://platform.quip.com" }, "disabled": false, "autoApprove": [] } } }

Utilization Methods

Upon successful service instantiation, the following procedural calls are exposed to the consuming AI model:

• quip_read_document: For fetching content via thread identifier.
• quip_append_content: For adding data to the tail of the document.
• quip_prepend_content: For adding data to the head of the document.
• quip_replace_content: For overwriting the entire document body.
• quip_create_document: Document provisioning capability (currently unimplemented).

Illustrative invocation sequence within an AI context:

quip quip_read_document { "threadId": "YOUR_DOCUMENT_ID" }

Backend Script Interface

The server logic delegates the actual API interaction to a specific Python utility located via the PYTHON_SCRIPT_PATH constant. This script must support the following functional modes:

• read: Document content retrieval.
• append: Tail-end data addition.
• prepend: Head-end data addition.
• replace: Full content substitution.

Licensing

Issued under the terms of the ISC License.

Originator

AvinashBole