In the video workflow, agents know to look in specific folders (video ingest) for raw files, process them through transcription, audio analysis, and storyboard generation, place outputs in labeled folders (motion graphics), enabling accurate multi-step processing without requiring agents to discover steps.

The three-layer pattern ensures that as raw data grows, deterministic scripts can extract needed information from the schema because the structure tells the agent exactly where to find specific data (like a client's audit or Remotion templates) without semantic search.

At APG Software, Adam has built seven different plugins using AI that drop project delivery time by 40-80% while enabling tasks like AI-based video editing, presentation building, and full day-to-day process automation.

An AI first business system or AI operating system is fundamentally a way of building different departments in a business using Claude code and organized agent plugins.

The Karpathy LLM Wiki framework involves a three-level structure: raw data ingestion, summarization and structuring of data, and schema creation that tells the AI where data is stored and how to access it, enabling knowledge base growth without loss of maintainability.

The audit extraction workflow connects directly to meeting transcripts (via Fathom.ai), client communications (via Twilio), and emails (via Gmail) to pull all contextual information into a single raw data layer before structuring it.

The audit JSON and related Wiki data are stored in a central knowledge base alongside Claude MD files and skill.md files that tell the agent what plugins, skills, data, and connectors exist in the system, enabling agent awareness of available tools.

The core objective of the three-layer architecture is controlled context injection into LLM conversations, which prevents the AI from indexing all raw data and bloating the context window when searching for information.

The video editor pipeline at APG Software uses Remotion and a three-layer structure similar to the audit system: raw video ingestion, storyboard JSON generation, and Remotion skill files that define pacing rules and templates for long-form, short-form, sales letters, and advertisements.

The video editor's ingest layer executes commands to transcribe video, create proxies (converting 4K to smaller files for easier API handling), perform audio analysis, transcription, and video clipping to remove dead noise—all happening before structured data generation.

In the video editor, the storyboard is a JSON structure that specifies for each clip: which motion graphic to show, the transcript text, zoom effects, etc., functioning as a Wiki layer that Remotion then reads and renders.

Claude is building agent memory functionality into its products, and Adam's team uses separate running memory markdown files that Claude updates, which is more effective and lightweight than RAG because it avoids the overhead of maintaining vector databases.

Understanding the three-layer structure and how to organize a repository folder is really important when building larger production systems, and this determines whether agents can efficiently route to information or must do expensive discovery.

At APG Software's video plugin, the code plugin folder structure corresponds directly to the video production folder structure, with raw video ingestion, processing, and output stages organized so agents know exactly which folders to access for each step.

Karpathy's LLM Wiki gained over 5,000 GitHub stars in 7 days and people are using it to build personal knowledge base systems on tools like Obsidian, but no one has shown how to operate this in a business context.

Without proper folder organization and clear labeling, agents perform extra digging to locate information, pulling unnecessary extra data into the context window, which reduces accuracy of responses.

Adam offers free templates and community resources through his school for learning how to build AI plugins and first systems, which he is showcasing in coming weeks with demonstrations of how they reduce time on various day-to-day activities.

At APG Software, the business audit process—which previously took 200+ hours—now takes 20-40 hours (an 80% reduction) because they implemented an AI first business system that automates the heavy lifting of data collection and organization.

The audit extractor agent pulls all connected data and stores it in a raw folder structure (e.g., meetings/transcripts), then extracts that data into JSON format to create the Wiki layer, which captures who completes tasks, duration, frequency, tools used, pain points, and proposed optimizations.

The core approach is just data organization using proven formatting like YAML and JSON to store information in a controlled way that enables scripts to filter information and extract exactly what's needed, preventing unnecessary context bloat from being injected into conversations.