All Your Base64

Provides an MCP server with encode/decode, memory, recall, and datauri tooling for AI workflows.

typescript

2

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "skullzarmy-all-your-base64": {
      "command": "ayb64",
      "args": [
        "mcp"
      ]
    }
  }
}

You can run All Your Base64 as an MCP server to empower AI workflows with memory, job recall, and quick encoding/decoding tools. The MCP server exposes a set of actions you can call from AI agents or code editors, making it easy to encode data, retrieve past results, and generate web-friendly data URIs within automated pipelines and development environments.

How to use

To use the MCP server, start it from your preferred entry point and connect your MCP client with a simple configuration. The server exposes these core tools: encode, decode, recall, jobs, datauri, memory. Start a session, queue a task, and retrieve the full result along with metadata for auditing and reproducibility. For typical workflows, you start the server and then issue tool calls like encode or datauri from your client, receiving structured results that include data, metadata, and timing information.

How to install

Prerequisites: you need Node.js installed on your system. Ensure you have npm available to install CLI tooling and run the MCP server commands.

Choose one of the MCP server start methods shown below and run the exact commands.

Option 1: Install globally and start the MCP server via the CLI command.

Install and start (concrete steps)

# Global installation (recommended for quick start)
npm install -g ayb64

# Start the MCP server using the CLI (recommended)
ayb64 mcp

Alternative start methods

# Start via direct binary
ayb64-mcp

# Development mode (if you are actively developing)
npm run mcp

# Or via TypeScript entry (source-driven)
npx tsx src/mcp-server.ts

Client configuration (example)

Configure your MCP client to connect to the local STDIO MCP server by specifying the command and arguments. If you use the ayb64 command, point the client to run ayb64 with the mcp subcommand.

Available tools

encode

Encode input data to base64 with support for formats and job tracking within the MCP workflow.

decode

Decode base64 input back to the original data with auto-detection and file output.

recall

Retrieve a previous job by ID, restoring full context and metadata for auditability.

jobs

List recent encoding/decoding operations with associated metadata.

datauri

Generate web-ready data URIs from input files for immediate web use.

memory

Manage the in-memory job cache, including stats and cache control.