Step into the Agentic Era

Bridge the Chasm.
Local Git State Meets Cloud AI.

A secure, metadata-only filesystem bridge that lets your cloud-based AI assistant "see" the exact state of your local repositories safely—preserving 100% of your raw source code privacy.

Get Started Free Explore Built-in MCP
terminal — digital_corpus_callosum.py
~ $ python3 digital_corpus_callosum.py --scan-dirs ~/Projects --output ./registry.json
Digital Corpus Callosum™ - Launching Scan...
Start time: 2026-06-03 10:00:15
Scanning directory: /Users/highpower/Projects
Found repository: PlurisUnum
Found repository: LifeOS
Scan complete! Registered 2 repositories successfully.
Registry successfully written to: ./registry.json

Engineered for Modern Workflows

A lean, secure implementation built entirely on standards to support the agentic developer toolchain.

1

Zero Dependencies

Built 100% on the standard Python library. Runs instantly without requiring pip installs or binary compilation.

2

Privacy First

Only parses local Git metadata and directory states. No raw codebase is uploaded, keeping proprietary IP fully on-device.

3

Built-In MCP Server

Runs native JSON-RPC 2.0 stdio server, serving repository state directly to agentic environments like Cursor or Claude Desktop.

4

Strict Schema Standard

Outputs registries following a strict JSON Schema, allowing any compliant pipeline to safely validate outputs.

Step 3 Specification

Standardized JSON Schema

Digital Corpus Callosum™ guarantees output stability by standardizing its repository registry schemas. Other tools, pipelines, or cloud agents can reliably run schema validation to ensure data consistency before reasoning over metadata.

dcc --print-schema
Prints the standardized JSON Schema description directly on your terminal.
schema.json JSON Schema
{
  "$schema": "http://json-schema.org/draft-07/schema#",
  "title": "Digital Corpus Callosum™ Registry Schema",
  "type": "object",
  "required": [
    "agent_name",
    "last_updated",
    "scanned_roots",
    "repository_count",
    "repositories"
  ],
  "properties": {
    "agent_name": { "type": "string" },
    "last_updated": { "type": "string" },
    "repository_count": { "type": "integer" }
  }
}
claude_desktop_config.json Configuration
{
  "mcpServers": {
    "digital-corpus-callosum": {
      "command": "python3",
      "args": [
        "/path/to/digital_corpus_callosum.py",
        "--mcp"
      ]
    }
  }
}
Step 4 Protocol

Native MCP Integration

Initiate an active local-to-cloud server with `--mcp`. Communicating via JSON-RPC 2.0 over standard input and output (stdio), DCC lets any compatible agentic coding environment trigger filesystem scans and inspect repository health directly.

get_repository_registry: Read-only tool that securely serves the current cached repository statuses.

scan_repositories_now: On-demand triggers an active recursive directory scan, updating the Drive synced registry.

Simple, Predictable Plans

Bring your own AI API keys. We focus on building and maintaining secure filesystem integration infrastructure.

Open Core

$0 MIT Licensed
  • Standard Python CLI
  • Stdio MCP Server
  • Schema Validation Support
Most Popular

DCC Pro

$8 / month
  • All Open Core features
  • Packaged Native Desktop App
  • Real-Time Filesystem Watchers
  • On-Device Local AI Diffs

Enterprise

$20 / seat / month
  • All Pro features
  • Shared Team configurations
  • Centralized compliance audits
  • Data Leakage Prevention rules