index 446a580..120cbd5 100644
@@ -1,3 +1,50 @@
---
visibility: public
\ No newline at end of file
+---
+
+# Tools & Implementations
+
+GitHub repos implementing the [[The Pattern|Karpathy LLM Wiki pattern]], ranked by stars (as of 2026-04-10).
+
+## Top Tier (1000+ stars)
+
+| Stars | Repo | Type | Notes |
+|------:|------|------|-------|
+| **1,542** | [SamurAIGPT/llm-wiki-agent](https://github.com/SamurAIGPT/llm-wiki-agent) | Multi-agent CLI | Works with Claude Code, Codex, Gemini CLI. No API key needed. **Community leader.** |
+
+## Mid Tier (300-500 stars)
+
+| Stars | Repo | Type | Notes |
+|------:|------|------|-------|
+| 481 | [nashsu/llm_wiki](https://github.com/nashsu/llm_wiki) | Desktop app | Cross-platform, docs to wiki |
+| 477 | [sdyckjq-lab/llm-wiki-skill](https://github.com/sdyckjq-lab/llm-wiki-skill) | Claude Code skill | Chinese community, multi-platform |
+| 347 | [AgriciDaniel/claude-obsidian](https://github.com/AgriciDaniel/claude-obsidian) | Obsidian companion | Claude + Obsidian integration |
+| 341 | [xoai/sage-wiki](https://github.com/xoai/sage-wiki) | MCP server | Captures knowledge from AI conversations |
+| 306 | [atomicmemory/llm-wiki-compiler](https://github.com/atomicmemory/llm-wiki-compiler) | Compiler | Raw sources in, interlinked wiki out |
+
+## Growing (100-300 stars)
+
+| Stars | Repo | Type | Notes |
+|------:|------|------|-------|
+| 264 | [Ar9av/obsidian-wiki](https://github.com/Ar9av/obsidian-wiki) | Framework | AI agents build/maintain Obsidian wiki |
+| 209 | [Astro-Han/karpathy-llm-wiki](https://github.com/Astro-Han/karpathy-llm-wiki) | Skill | One skill to build your own Karpathy-style LLM wiki |
+| 192 | [lucasastorian/llmwiki](https://github.com/lucasastorian/llmwiki) | MCP server | Upload docs, connect Claude, it writes your wiki |
+| 159 | [nvk/llm-wiki](https://github.com/nvk/llm-wiki) | Multi-agent | Research, ingestion, compilation, querying |
+| 147 | [ussumant/llm-wiki-compiler](https://github.com/ussumant/llm-wiki-compiler) | Claude Code plugin | Compiles markdown, ~90% context reduction |
+
+## Early Stage (< 100 stars)
+
+| Stars | Repo | Type |
+|------:|------|------|
+| 77 | [Ss1024sS/LLM-wiki](https://github.com/Ss1024sS/LLM-wiki) | Baseline implementation |
+| 51 | [kfchou/wiki-skills](https://github.com/kfchou/wiki-skills) | Claude Code skills |
+| 46 | [Pratiyush/llm-wiki](https://github.com/Pratiyush/llm-wiki) | Multi-platform |
+| 29 | [MehmetGoekce/llm-wiki](https://github.com/MehmetGoekce/llm-wiki) | L1/L2 cache + Logseq |
+| 28 | [iamsashank09/llm-wiki-kit](https://github.com/iamsashank09/llm-wiki-kit) | MCP server on Glama |
+| 28 | [toolboxmd/karpathy-wiki](https://github.com/toolboxmd/karpathy-wiki) | Claude Code skills |
+| 27 | [ekadetov/llm-wiki](https://github.com/ekadetov/llm-wiki) | Obsidian-backed plugin |
+
+## Packages
+
+- **npm:** `llm-wiki` — CLI with multi-provider support (OpenAI, Anthropic, DeepSeek, Ollama), auto-linking, wiki linting
+- **PyPI:** `llmwiki` — Python package, no vector DB or embeddings needed
\ No newline at end of file