Skip to main content
Glama

delete_blob

Remove a blob from Walrus decentralized storage by specifying its ID. This tool supports efficient data management and storage optimization on the blockchain-verified Sui network.

Instructions

Delete a blob from Walrus storage

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
blobIdYesThe blob ID to delete

Implementation Reference

  • MCP tool handler for 'delete_blob' that validates input with DeleteBlobSchema and delegates to walrusClient.deleteBlob, returning the result as JSON text.
    case 'delete_blob': { const { blobId } = DeleteBlobSchema.parse(args); const result = await walrusClient.deleteBlob(blobId); return { content: [ { type: 'text', text: JSON.stringify(result, null, 2), }, ], };
  • Zod input schema for the delete_blob tool, defining the expected 'blobId' parameter.
    const DeleteBlobSchema = z.object({ blobId: z.string().describe('The blob ID to delete'), });
  • src/index.ts:102-115 (registration)
    Registration of the 'delete_blob' tool in the ListToolsRequestHandler, providing name, description, and input schema.
    { name: 'delete_blob', description: 'Delete a blob from Walrus storage', inputSchema: { type: 'object', properties: { blobId: { type: 'string', description: 'The blob ID to delete', }, }, required: ['blobId'], }, },
  • WalrusClient.deleteBlob method called by the tool handler. Currently returns a message explaining that manual deletion is not supported in Walrus; blobs expire automatically.
    async deleteBlob(blobId: string): Promise<{ success: boolean; message: string }> { // Note: Walrus blobs cannot be deleted once stored - they expire based on their epoch settings return { success: false, message: 'Walrus blobs cannot be manually deleted. They will expire automatically based on their storage epochs.', }; }

Other Tools

Related Tools

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/Mr-Sunglasses/walrus-mcp'

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