digital-brain

A personal knowledge system for founders coordinates AI across voice, brand, content, network, and operations. It uses progressive disclosure to load only relevant modules for tasks like content creation, CRM, and weekly reviews, minimizing context load. Data is stored in append-only JSONL, YAML, and MD formats.

Append-only memory stores all ideas, contacts, interactions, and posts in versioned JSONL, YAML, and MD files. This preserves history and supports retrospectives by never overwriting past entries, allowing AI to reason over a complete, chronological log of your personal and operational data.

You can automate routine workflows using self-contained Python scripts. These scripts handle tasks like weekly reviews, content idea generation, stale-contacts checks, and idea-to-draft expansion, operating directly on your stored JSONL and YAML data without requiring external dependencies.

Yes, by defining your voice in identity/voice.md and positioning in identity/brand.md, the system can draft content via templates. It logs the generated outputs for auditing, ensuring the content aligns with your established brand identity and personal voice.

No, the system operates with zero external dependencies. It relies on self-contained Python scripts for automation and stores all data locally in JSONL, YAML, and Markdown files, ensuring module isolation and progressive disclosure without needing external APIs.

Progressive disclosure is a mechanism that loads only the relevant module—such as identity, content, network, or operations—for the current task. This approach minimizes context load and optimizes AI reasoning by preventing the model from processing unnecessary information.