AUTOMATION

Bootstrap a custom knowledge-base system with a single agent seed prompt

Use LLM agents to auto-generate markdown-to-HTML knowledge systems with flexible customization

Updated: 5/22/2026
Difficulty
medium
Time
30m
Use Case
Create a personal, customizable knowledge-base SaaS from a single seed prompt that converts markdown folders into browsable HTML and supports agent interrogation
Popularity
0 views

About this automation

A seed prompt system that instructs an AI agent to build a complete knowledge-base infrastructure: markdown folder organization, Python script generation for HTML conversion, and agent-queryable interface. The system is flexible and can be customized by natural language instructions. Designed to prevent context pollution by keeping knowledge systems separate from coding projects.

How to implement

1

Create a fresh empty directory (do not use existing project folders)

2

Paste the seed prompt into Claude Code or Cowork

3

Let the agent onboard you through Phase 1-2 setup

4

Agent builds Python script to convert markdown to static HTML site

5

Agent requests permission before file operations (Phase 3)

6

Review and customize the system by telling the agent what you want

7

Use the generated HTML interface to browse knowledge base

8

Query and update knowledge base via agent interaction

9

Clone coding projects into the folder later if needed (after setup)