Skip to main content
Glama

checkpoint_create

Save a snapshot of your current workspace state to preserve progress and enable rollback capabilities.

Instructions

Create a checkpoint of the current workspace state.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
requestYes

Implementation Reference

  • The checkpoint_create method in CheckpointTools that executes the tool logic: creates a checkpoint via checkpoint_manager and returns a response.
    async def checkpoint_create( self, request: CheckpointCreateRequest ) -> CheckpointCreateResponse: """ Create a checkpoint of the current workspace state. Args: request: CheckpointCreateRequest with name and description Returns: CheckpointCreateResponse with checkpoint info """ checkpoint_info = await self.checkpoint_manager.create_checkpoint( name=request.name, description=request.description ) return CheckpointCreateResponse( success=True, checkpoint_id=checkpoint_info.id, created_at=checkpoint_info.created_at, )
  • Registers the checkpoint_create tool on the MCP server using @server.tool decorator, delegating to the CheckpointTools instance.
    async def checkpoint_create(request: CheckpointCreateRequest): """Create a checkpoint of the current workspace state.""" return await checkpoint_tools_instance.checkpoint_create(request)
  • Input schema: Pydantic model CheckpointCreateRequest defining name (optional) and description fields.
    class CheckpointCreateRequest(BaseModel): """Request to create checkpoint""" name: str | None = None description: str = ""
  • Output schema: Pydantic model CheckpointCreateResponse with success, checkpoint_id, and created_at.
    class CheckpointCreateResponse(BaseModel): """Response from checkpoint creation""" success: bool checkpoint_id: str created_at: datetime
  • Instantiates the CheckpointTools instance with checkpoint_manager, used for tool delegation.
    checkpoint_tools_instance = CheckpointTools(checkpoint_manager)

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/chrishayuk/chuk-mcp-vfs'

If you have feedback or need assistance with the MCP directory API, please join our Discord server