Distill KB-worthy knowledge from the current conversation — identifies concepts, connections, and insights worth preserving, then creates or updates notes in the vault. Use whenever the user says 'distill the KB', 'extract knowledge from this conversation', 'what did we learn', or invokes /kb-distill or $kb-distill. Also triggers for 'save what we figured out', 'add our findings to KB', or 'capture these insights'.
Check knowledge base vault consistency — finds orphan notes and oversized articles. Use whenever the user says 'check KB health', 'audit the vault', or invokes /kb-health. Also triggers for 'are there any KB issues', 'vault consistency', or 'find orphan notes'.
Ingest a source (PDF, Zotero item, code repository, or URL) into the Obsidian knowledge base — creates a curated article with key concepts and references. Use whenever the user wants to add something to the KB, says 'ingest this', 'add this paper/repo/PDF/article to KB', 'index this', or invokes /kb-ingest or $kb-ingest. Also triggers for 'create a KB article from this', 'add this to my knowledge base', or any mention of importing academic papers, codebases, web articles, documentation, or documents into the vault.
Refine and deepen an existing knowledge base article by re-reading raw sources with a specific question in mind. Use when gap detection identified missing knowledge, or when the agent needs to deepen an article that was too shallow for the current task. Triggers for 'deepen this article', 'refine the KB on X', 'fill in this gap', 'the KB was too shallow on X'.
Semantic search over the user's knowledge base (Obsidian vault). Invoke BEFORE any factual lookup — before WebFetch, before WebSearch, before reading source files in cloned repos, before grep'ing a codebase for documentation, before answering 'how does X work' or 'what is X' from training data. Also invoke before writing new KB content (kb-ingest, kb-distill) to avoid duplication. Trigger phrases: 'search the KB', 'what do I have on X', 'find related notes', 'check my notes'. The KB is the authoritative source for any topic the user has previously ingested — assume notes may exist and check first.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A collection of Agent Skill plugins for Claude Code and Codex.
| Plugin | Description |
|---|---|
| obsidian-kb | Knowledge base management via Obsidian vault — ingest sources, distill conversation insights, refine articles, check vault health |
| lean4-skills-lite | Lightweight Lean 4 skills — library search, proof assistance |
npx claudepluginhub szch79/agent-marketplace --plugin obsidian-kbLightweight Lean 4 skills — library search, proof assistance
Persistent, compounding knowledge base maintained by LLMs in Obsidian — agent-first edition. Four task-oriented agents (Researcher / Advisor / Curator / Scribe) with citations, confidence, supersession, and rolling session cache. Inspired by Karpathy's LLM Wiki pattern.
Bidirectional knowledge flow between Claude Code and Obsidian — 20 MCP tools, skills, and hooks for PKM
Second Brain automation for Obsidian vaults — entity management, ingestion, compression, and sync via Claude Code skills
LLM Wiki — persistent, compounding knowledge base inside Obsidian. Ingest, compile, query, lint, and remove knowledge base wikis. Based on Karpathy's LLM Wiki pattern.
Build and maintain LLM-powered knowledge bases as Obsidian wikis with compile, query, lint, and evolve workflows
Kit completo de skills pra transformar uma pasta em vault Obsidian profissional + dashboard local com ML. 6 skills: init (scaffold), librarian (cura contínua), migrate (adoção não-destrutiva), organizer (clusters/duplicatas), expand (notas-ponte via LLM restrito), pulse (dashboard localhost com FSRS, recommendations, anomalies). 100% local, privacy-first, pt-br.