llm-wiki

A pattern for building an AI-powered Obsidian wiki, where knowledge is distilled into linked Markdown notes. It separates original sources, collected material, and maintained notes so the resulting knowledge base is easier to search and update.

In plain words
What is it for?
Use it to understand or set up the wiki’s three-layer structure, ingest documents and images, and maintain a compiled knowledge base. Image interpretation requires a model that can see images.
Why use it?
It prevents the system from repeatedly rediscovering the same information and keeps original documents unchanged as references.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/ar9av/obsidian-wiki/llm-wiki
Any agent
npx skills add Ar9av/obsidian-wiki --skill llm-wiki
Clone the repo
git clone --depth 1 https://github.com/Ar9av/obsidian-wiki

Made for: Claude Code, Codex.

Per session 113 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 9,009 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00113 $0.09009
Opus 5 $0.00056 $0.04505
Sonnet 5 $0.00023 $0.01802
Haiku 4.5 $0.00011 $0.00901

Measured today against content hash e0e01f88e1a6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

llm-wiki scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.skills/llm-wiki/SKILL.md · 657 lines

How it starts

The opening of the file, as written. The whole thing — 657 lines — stays where its author put it; the contents beside it link to each section on GitHub.

LLM Wiki — Knowledge Distillation Pattern

You are maintaining a persistent, compounding knowledge base. The wiki is not a chatbot — it is a compiled artifact where knowledge is distilled once and kept current, not re-derived on every query.

Three-Layer Architecture

Layer 1: Raw Sources (immutable)

The user's original documents — articles, papers, notes, PDFs, conversation logs, bookmarks, and images (screenshots, whiteboard photos, diagrams, slide captures). These are never modified by the system. They live wherever the user keeps them (configured via OBSIDIAN_SOURCES_DIR in .env). Images are first-class sources: the ingest skills read them via the Read tool's vision support and treat their interpreted content as inferred unless it's verbatim transcribed text. Image ingestion requires a vision-capable model — models without vision support should skip image sources and report which files were skipped.

Think of raw sources as the "source code" — authoritative but hard to query directly.

Don't confuse this with the in-vault _raw/ staging folder, which is a different thing: a scratch inbox for quick captures and drafts awaiting promotion (see wiki-capture and wiki-ingest). Files there aren't Layer 1 sources, but wiki-ingest still moves rather than deletes them on promotion, since some have no other copy.

Layer 2: The Wiki (LLM-maintained)

A collection of interconnected Obsidian-compatible markdown files organized by category. This is the compiled knowledge — synthesized, cross-referenced, and navigable. Each page has:

  • YAML frontmatter (title, category, tags, sources, timestamps)
  • Obsidian [[wikilinks]] connecting related concepts
  • Clear provenance — every claim traces back to a source

The wiki lives at the path configured via OBSIDIAN_VAULT_PATH in .env.

Layer 3: The Schema (this skill + config)

The rules governing how the wiki is structured — categories, conventions, page templates, and operational workflows. The schema tells the LLM how to maintain the wiki.

Read the full file on GitHub · 657 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. today Changed · +4 lines e0e01f88e1a6
  2. yesterday First seen · 653 lines · 113 tokens per session scan A 2debf9365116

Subscribe to this mod's changes

llm-wiki is a skill published in the GitHub repository Ar9av/obsidian-wiki (3,325 stars, last pushed today), licensed MIT. It adds 113 tokens to every session and 9,009 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

deeprefine

Agent-native DeepRefine refinement loop — same control flow as DeepRefine.refine(), graphify search instead of FAISS, session LLM, dry-run review before approved graph writes.

HKUST-KnowComp/DeepRefine-Skill · 42 tokens

deeprefine

Agent-native DeepRefine refinement loop — same control flow as DeepRefine.refine(), graphify search instead of FAISS, session LLM, dry-run review before approved graph writes.

HKUST-KnowComp/DeepRefine-Skill · 42 tokens

deeprefine

Claude Code adapter for the DeepRefine agent-native refinement loop. Use when the user invokes /deeprefine, or asks to refine, diagnose, review, or apply changes to a Graphify / LLM-Wiki knowledge graph. Must follow the canonical DeepRefine skill rules and stop for review before graph writes.

HKUST-KnowComp/DeepRefine-Skill · 68 tokens

binder-modeling

Binder data modeling — define entity types, fields, relations, constraints, views, and navigation. Use when asked to "create a type", "add a field", "define a schema", "set up relations", "model entities", "create a view", "set up navigation", "render entities as files", or design a binder workspace schema.

mpazik/Binder · 74 tokens

binder-app

Build a small HTTP app (dashboard, kanban board, admin panel, custom UI) on top of an existing Binder workspace. Use when asked to "create an app", "build a dashboard", "show records in a chart", "make a kanban/board", "build an admin panel", or "add a UI on top of binder".

mpazik/Binder · 74 tokens

binder-cli

Binder CLI for knowledge graph operations — CRUD, search, schema inspection, transaction import, docs rendering. Use when asked to "query binder", "search records", "create a record", "check the schema", "import transactions", "undo changes", or work with a binder workspace.

mpazik/Binder · 60 tokens