Follow these steps to launch the MCP Server and use its tools with a chatbot in a temporary Meshagent Room for testing:
  1. Install Meshagent 
    pip install "meshagent[all]"
  2. Sign Up and Authenticate
  3. Start an MCP Server in a Room 
    meshagent service test --room=test --role=agent --image=meshagent/mcp-basic-memory:latest --env MESHAGENT_PORT=8001 --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" --name=mcp-basic-memory-service-test
    • This creates a Meshagent Room named test and launches an MCP server in it. The room and all services will shut down automatically when inactive.
  4. Add a Chatbot Connected to the MCP Toolkit 
    meshagent chatbot join --room=test --agent-name=mcp-basic-memory --name=mcp-basic-memory --toolkit=mcp-basic-memory
    • This command launches a chatbot in the same Room, connected to the MCP Basic Memory toolkit.
    • The console output will include a link to the Room.
  5. Interact via Browser
    • Visit the provided Room link to chat with the agent and use MCP Server tools directly!

​
Project Level Deployment (Production Setup)

You can configure services to start automatically in every new Room created under your Meshagent project—ideal for persistent, always-available AI and tool access.
  1. Create a Persistent MCP Server Service 
    meshagent service create --role=agent --image=meshagent/mcp-basic-memory:latest --env MESHAGENT_PORT=8001 --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" --name=mcp-basic-memory-service
  2. Create a Persistent Chatbot Service with MCP Toolkit 
    meshagent service create --image="meshagent/cli:latest" --port="num=9001 path=/agent liveness=/ type=meshagent.callable participant_name=mcp-basic-memory-chatbot" --env="MESHAGENT_PORT=9001" --name="mcp-basic-memory-chatbot-service" --command="meshagent chatbot service --agent-name=mcp-basic-memory-chatbot --toolkit=mcp-basic-memory"
Now, every Room created in your project will automatically load both the MCP server and a chatbot connected to its toolkit. No need to run processes manually; they join every Room on creation.

​
Tools Available

​
Tools Provided by MCP Basic Memory Server

ToolShort Description
build_contextBuild context from a memory:// URI to continue conversations naturally.
canvasCreate an Obsidian canvas file to visualize concepts and connections.
delete_noteDelete a note by title or permalink.
project_infoGet information and statistics about the current Basic Memory project.
read_contentRead a file’s raw content by path or permalink.
read_noteRead a markdown note by title or permalink.
recent_activityGet recent activity from across the knowledge base.
search_notesSearch across all content in the knowledge base.
write_noteCreate or update a markdown note.

​
Tools Details

​
build_context

Build context from a memory:// URI to continue conversations naturally.
  • Use this to follow up on previous discussions or explore related topics.
  • Timeframes support natural language (e.g., "2 days ago", "last week", "today", "3 months ago") or standard formats ("7d", "24h").
Parameters:
NameTypeDescription
urlstring
depthstringoptional
max_relatedintegeroptional
pageintegeroptional
page_sizeintegeroptional
timeframestringoptional

​
canvas

Create an Obsidian canvas file to visualize concepts and connections. Parameters:
NameTypeDescription
edgesarray
folderstring
nodesarray
titlestring

​
delete_note

Delete a note by title or permalink. Parameters:
NameTypeDescription
identifierstring

​
project_info

Get information and statistics about the current Basic Memory project.

​
read_content

Read a file’s raw content by path or permalink. Parameters:
NameTypeDescription
pathstring

​
read_note

Read a markdown note by title or permalink. Parameters:
NameTypeDescription
identifierstring
pageintegeroptional
page_sizeintegeroptional

​
recent_activity

Get recent activity from across the knowledge base.
  • Timeframe supports natural language formats (e.g., "2 days ago", "last week", "yesterday", "today") or standard formats ("7d").
Parameters:
NameTypeDescription
depthintegeroptional
max_relatedintegeroptional
pageintegeroptional
page_sizeintegeroptional
timeframestringoptional
typestringoptional

​
search_notes

Search across all content in the knowledge base. Parameters:
NameTypeDescription
querystring
after_datestringoptional
entity_typesstringoptional
pageintegeroptional
page_sizeintegeroptional
search_typestringoptional
typesstringoptional

​
write_note

Create or update a markdown note. Returns a markdown formatted summary of semantic content. Parameters:
NameTypeDescription
contentstring
folderstring
titlestring
tagsstringoptional