Use when building, querying, revising, or linting a persistent LLM-maintained wiki from source material in Dreadnode Agent Output.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# LLM Wiki
Compile source material into `wiki_page` records that compound across turns.
Operate on the maintained pages instead of repeatedly reconstructing knowledge
from raw sources.
This instantiates Andrej Karpathy's
[LLM Wiki](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f)
pattern with typed Agent Output records rather than Markdown files.
Choose one stable lowercase kebab-case `wiki_id` before reading or writing. Put
that value on every page and pass it to `wiki_progress`; it is the isolation
boundary when a project contains more than one wiki.
## Choose the operation
- **Ingest** when the user supplies new source material.
- **Query** when the user asks a question about accumulated knowledge.
- **Lint** when the user asks to check or improve wiki health.
## Ingest
1. Call `list_items` with `item_type="wiki_page"` (or `search_items` with a
topic query and `item_type="wiki_page"`) to find existing pages related to
the source.
2. Read only those pages with `read_item`. Reject candidates whose `wiki_id`
differs from the active wiki.
3. Read the raw source and extract claims with concrete evidence references.
…
The line at the top of this page pins the label git:20260728.71566e8, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
GET https://markdownregistry.com/api/v1/artifacts/art_me4266et7kklkckc
GET https://markdownregistry.com/api/v1/resolve?ref=dreadnode/capabilities/llm-wiki
GET https://markdownregistry.com/api/v1/blob/7f32a2b751842950ce0d8d35530bbcb8ca4bdd7e22223c2aad88b4b3dea2c3e1
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
dreadnode/capabilities · capabilities/ai-red-teaming/skills/analytics-interpretation/SKILL.md · Guide for reading and interpreting AIRT analytics output including ASR, risk scores, severity, and compliance metrics
dreadnode/capabilities · capabilities/ai-red-teaming/skills/attack-selection-guide/SKILL.md · Decision tree for selecting AIRT attacks based on goals, target type, and constraints
dreadnode/capabilities · capabilities/ai-red-teaming/skills/auth-setup-guide/SKILL.md · Guide the user through authenticating the target, attacker, and judge models from their own environment (any cloud, any…
dreadnode/capabilities · capabilities/ai-red-teaming/skills/compliance-mapping/SKILL.md · Maps AIRT attacks, transforms, and scorers to OWASP LLM Top 10, OWASP ASI (ASI01-ASI10), MITRE ATLAS, NIST AI RMF, and…
dreadnode/capabilities · capabilities/ai-red-teaming/skills/error-troubleshooting/SKILL.md · Diagnose and fix common errors in AIRT attack generation, execution, and results collection
dreadnode/capabilities · capabilities/ai-red-teaming/skills/provisioning-and-lifecycle/SKILL.md · Provision a hosted target, pick the right endpoint per target type, tear down to stop billing, and interpret transient…
dreadnode/capabilities · capabilities/ai-red-teaming/skills/transform-reference/SKILL.md · Catalog of AIRT transforms - 429 text transforms (596 incl. multimodal/variants); documents the core set with exact…
dreadnode/capabilities · capabilities/ai-red-teaming/skills/whitebox-local-surrogate/SKILL.md · How to run white-box (gradient/embedding/activation) attacks, which need a local open-weight surrogate model the user…
ar9av/obsidian-wiki · skills/llm-wiki/SKILL.md · Explain and apply the foundational LLM Wiki architecture and knowledge-management strategy. Use for setup…
cowork-os/cowork-os · resources/skills/llm-wiki/SKILL.md · Build and maintain a workspace-local, Obsidian-friendly research vault using Karpathy's LLM Wiki pattern.
jellydn/my-ai-tools · configs/amp/plugins/my-ai-tools-skills/skills/llm-wiki/SKILL.md · Build a persistent compounding wiki from raw sources using the LLM Wiki pattern
infranodus/skills · skill-llm-wiki/SKILL.md · Guide users through setting up a personal LLM-maintained wiki — a persistent, compounding knowledge base where the LLM…
praneybehl/llm-wiki-plugin · skills/llm-wiki/SKILL.md · Build and maintain an LLM-curated knowledge base from papers, articles, transcripts, notes and project findings. Ingest…
seaworld008/commonly-used-high-value-skills · openclaw-skills/llm-wiki/SKILL.md · Use when building, querying, ingesting sources into, or auditing an interlinked Markdown knowledge base with immutable…
