Browse / LLM wiki / llm-wiki

llm-wiki

joe12803

Based on Andrej Karpathy's LLM Wiki pattern.

Open joe12803/llm-wiki → Checked against the four rules.

Does not say

no agent config and no instruction file at the root.

In its own words

README.mdjoe12803/llm-wiki

Based on Andrej Karpathy's LLM Wiki pattern. A compounding, persistent, and organized knowledge base constructed as interlinked markdown files.

This repository tracks architectural diagrams, deployments, system environments, and conceptual guides for various tools integrated with our multi-agent framework (Hermes Agent).

📅 Chronological Logs

For daily synchronization records and system-wide steady-state verification history, check log.md.

🛠️ Main Projects & Entities

  • ds2api: DeepSeek-to-OpenAI API bridge middleware. Tracks ARM64 optimizations, Nginx proxy setups, SSE buffer tuning, and non-root Docker permissions.
  • gitnexus: AI-native code graph indexer and MCP server. Maps codebases to structured graphs for tools like Claude Code and Cursor.
  • hermes-claw: Deployment hub and session storage for Hermes Agent. Details credentials sync and persistent volume mapping.
  • claude-api-bridge: High-availability middleware bridging reverse-engineered Claude sessions.
  • evolver: Multi-agent system optimization and self-evolving loops.

Quoting 63% of the README. Read the rest on GitHub ↗

Read from the repository's git tree on 2026-08-17.