LucidLink MCP Server

The LucidLink MCP Server exposes filespace capabilities as tools that MCP-compatible agents can use directly. It gives agents and teams a shared, persistent data layer to read, write and work with enterprise data through a filespace they can use together — without moving the data first.

The server is distributed as the lucidlink-mcp package and is launched by your MCP client on demand, so there is no separate hosted service to deploy. It authenticates with a LucidLink service-account token — the same credential used across the LucidLink API and Python SDK.

The LucidLink MCP Server is in beta. Capabilities and command names may change between releases.

What agents can do

LucidLink MCP Server exposes filespace operations as tools that MCP-compatible agents can call directly:

Capability	Description
Read filespace data	Agents read briefs, notes, documents, transcripts, folders and project files directly from a LucidLink filespace.
Write outputs back	Agents create summaries, reports, checklists, briefs or extracted data and save them back as normal files.
Search the filespace	Agents find relevant files and content without relying on manually pasted or staged context.
Lock before editing	Agents coordinate edits using LucidLink file locking, including whole-file or byte-range advisory locks.
Audit file activity when enabled	When audit is enabled by an admin, audit data helps teams understand file activity such as who changed what and when.
Run read-only	Read-only mode lets agents inspect filespace data without being able to modify it.

Read filespace data

Description: Agents read briefs, notes, documents, transcripts, folders and project files directly from a LucidLink filespace.

Write outputs back

Description: Agents create summaries, reports, checklists, briefs or extracted data and save them back as normal files.

Search the filespace

Description: Agents find relevant files and content without relying on manually pasted or staged context.

Lock before editing

Description: Agents coordinate edits using LucidLink file locking, including whole-file or byte-range advisory locks.

Audit file activity when enabled

Description: When audit is enabled by an admin, audit data helps teams understand file activity such as who changed what and when.

Run read-only

Description: Read-only mode lets agents inspect filespace data without being able to modify it.

Under the hood these map to the filespace tools the server exposes — for example read_file, write_file, list_files, find_files, grep_files, edit_lines, lock_byte_range / unlock_byte_range, and read_audit_trail — the same filespace operations available through the Python SDK.

Get started in minutes

Step 1: Install and configure

Install uv, run the LucidLink MCP Server with uvx, then configure your service-account token.

# Install uvcurl -LsSf https://astral.sh/uv/install.sh | sh
# Verify the LucidLink MCP serveruvx lucidlink-mcp@latest --version
# Configure your LucidLink service-account tokenuvx --from lucidlink-mcp@latest lucidlink-mcp-setup

Step 2: Register with your MCP client

Claude

claude mcp add lucidlink -s user -- uvx lucidlink-mcp@latest

Codex

codex mcp add lucidlink -- uvx lucidlink-mcp@latest

Cursor

{  "mcpServers": {    "lucidlink": {      "command": "uvx",      "args": ["lucidlink-mcp@latest"]    }  }}

No infrastructure to deploy — your MCP client launches the server when needed. No separate hosted service is required.

Designed to run in your environment

Capability	Description
Runs where the agent runs	The MCP server is launched by the MCP client and does not require a separate hosted service.
No broad desktop mount required	Agents can work through MCP or the Python SDK without exposing decrypted files through a mounted drive.
Read-only mode	Agents can inspect filespace data without being able to modify it.
Dangerous-action confirmation	Dangerous or destructive operations are flagged for explicit confirmation by the client.
Service-account access	Use a LucidLink service-account token to give the MCP server an identity.
Audit trail support	Audit file and agent activity when audit is enabled by an admin.

Runs where the agent runs

Description: The MCP server is launched by the MCP client and does not require a separate hosted service.

No broad desktop mount required

Description: Agents can work through MCP or the Python SDK without exposing decrypted files through a mounted drive.

Read-only mode

Description: Agents can inspect filespace data without being able to modify it.

Dangerous-action confirmation

Description: Dangerous or destructive operations are flagged for explicit confirmation by the client.

Service-account access

Description: Use a LucidLink service-account token to give the MCP server an identity.

Audit trail support

Description: Audit file and agent activity when audit is enabled by an admin.

Resources

Get the package. Install or inspect the LucidLink MCP Server on PyPI.

Examples and workflows. Use the Python SDK, wrap Python SDK methods as agent tools, or connect agents through MCP with the agentic examples on GitHub.