Project Name
Vault template
First committed by asakin
Published in asakin/llm-context-base. Publisher page
74 GitHub stars
What it is
A git template for building your own LLM-powered personal wiki. Training period, metadata standard, lint system included. Clone and go.
AutoGineer overview, from the brain’s own files
llm-context-base is an opinionated template designed to turn a local git repository into an LLM-powered personal wiki. It implements a three-layer pattern consisting of raw source documents, an LLM-maintained wiki, and a schema configuration. The system is designed to adapt to a user's workflow over time, eliminating the need for upfront taxonomy design. It includes a metadata standard for queryable files, an inbox-first capture system with a 7-day TTL, and a linting system to maintain document health. It supports various AI tools including Claude Code, Cursor, and GitHub Copilot, providing a consistent interface for managing personal knowledge.
How it works
Memory model: LLM wiki
AutoGineer overview, from the brain’s own files
The system uses a specific folder structure including 1-Projects, 2-Knowledge, 3-Journal, and 4-Private, alongside configuration and metadata folders. Users capture information into an _inbox folder, which the AI then processes and files according to a metadata standard. The system is bootstrapped via configuration files in .claude, .cursor, and .github, which direct the AI to follow instructions located in _config and _meta. It uses a training period to learn user patterns and maintains documentation through a split README file that separates user-specific content from framework documentation.
| Folder | Role | Files |
|---|---|---|
| .claude | agent-config | 2 |
| .cursor | agent-config | 1 |
| .github | agent-config | 1 |
| .obsidian | obsidian-config | 7 |
| assets | attachments | 2 |
| docs | docs | 13 |
| examples | examples | 4 |
| 1-Projects | folder | 3 |
| 2-Knowledge | folder | 5 |
| 3-Journal | folder | 1 |
| 4-Private | folder | 1 |
| _config | folder | 4 |
| _inbox | folder | 1 |
| _meta | folder | 14 |
| _output | folder | 1 |
| _sources | folder | 2 |
The agent reads
- CONTRIBUTING.md evidence: .claude/CLAUDE.md
- _config/config.md evidence: README.md
The agent writes
- _config/config.md evidence: .claude/CLAUDE.md
- _config/context.md evidence: .claude/CLAUDE.md
- _inbox/ evidence: .claude/CLAUDE.md
- _output/ evidence: README.md
Requirements
Apps: Obsidian
Databases: MongoDB, PostgreSQL
Works with: Claude Code, GitHub Copilot, Cursor, Devin (ex-Windsurf)
Set it up
- Download the brain as a ZIP and unzip it where you keep your notes or project: https://mp-staging.autogineer.com/agent-brains/asakin/llm-context-base/download
In Obsidian, choose “Open folder as vault” and pick the folder.
Claude Code reads .claude/CLAUDE.md from this folder automatically.
cd llm-context-base && claude
AutoGineer scanned these files at commit 3c096ca.
First committed by asakin. Published in asakin/llm-context-base. Licence: Apache License 2.0.
Usage
AutoGineer overview, from the brain’s own files
- Capture unstructured notes into the inbox for later organization by the AI.
- Maintain decision records with context, options, and rationale.
- Query the wiki to synthesize information across multiple documents.
- Run linting commands to identify stale files or missing metadata.
- Manage project initiatives with briefs and timeline tracking.
- Generate reports or summaries from stored knowledge.
- “Capture this: here's how our onboarding process works...”
- Creates a file in the inbox and offers to file it in the appropriate location.
- “We decided to use PostgreSQL over MongoDB, let me document why”
- Generates a decision record with context, options, rationale, and an outcome section.
- “What do we know about our hiring process?”
- Scans file summaries and synthesizes an answer with source citations.
- “Lint the wiki”
- Surfaces stale files, missing metadata, and orphaned documents.
Documentation by asakin
[Project Name]
This README will be personalized during your first session.
<!-- llm-context-base:framework-readme -->
llm-context-base
An opinionated template for building your own LLM Wiki.
Use the template, point your AI assistant at it, and start talking. Over time, it learns how you work and adapts its structure to your needs without you having to design anything upfront.
How This Relates to Karpathy's LLM Wiki
Andrej Karpathy's LLM Wiki gist describes a three-layer pattern: raw source documents → an LLM-maintained wiki → a schema config file (like CLAUDE.md or AGENTS.md). The LLM reads sources, maintains structured wiki pages, and follows the schema config to stay consistent.
llm-context-base implements this pattern as a ready-to-use template — so you don't have to build it from scratch. What it adds on top of the base pattern:
Metadata standard — every file gets a queryable header (type, summary, tags, status, date). The LLM scans these instead of loading full files, which scales cleanly to hundreds of documents.
Inbox-first capture — a frictionless landing zone with a 7-day TTL. You capture without deciding where things go. The LLM files them later.
Training period — the system starts chatty (learning your patterns) and goes quiet as it adapts. You never design a taxonomy upfront.
Lint system — surfaces stale files, missing metadata, and orphaned docs on demand.
Multi-LLM support — bootstrap shims for Claude Code, Cursor, Copilot, Windsurf, and others. One config, any tool.
Full comparison with the Karpathy pattern →
Quick Start
1. Create your copy. Click "Use this template" on GitHub. Make it private. Clone it.
git clone https://github.com/YOUR-USERNAME/my-wiki.git
cd my-wiki2. Set up your workspace. Open the repo as an Obsidian vault on one side, your AI agent on the other. You talk to the AI — it builds the wiki. You watch it happen in Obsidian.
Full Obsidian setup → · No Obsidian? Any editor + any supported AI tool works fine.
3. Fill in your profile. Open _config/config.md, fill in the About You section, and set Training Start Date to today.
4. Start talking. Open in your AI tool of choice and go:
"Capture this: here's how our onboarding process works..."
"We decided to use PostgreSQL over MongoDB, let me document why"
"What do we know about our deployment process?"
"Lint the wiki, what needs attention?"
That's it. The system adapts to what you actually need.
Want to stay updated as the framework improves? The template approach gives you a clean start but disconnects you from upstream. If you want to pull future improvements (new protocols, better training behavior, bug fixes), use the Framework Track setup instead.
What Makes This Different
It learns how you work. The first 30 days are a training period. Your AI asks questions, suggests directories, and logs what it learns. By day 45, it's quiet and efficient. You never design a taxonomy upfront. Full details →
Capture first, organize later. Everything lands in _inbox/ with a 7-day TTL. You never hesitate to capture because you don't have to decide where it goes yet. Your AI files it later.
Your README stays yours. This file has two sections split by a marker. Above it: your personal project README, maintained by your AI as the wiki evolves. Below it: the framework docs, which track upstream cleanly. You get living documentation without merge conflicts.
Patterns are captured as they emerge. When your AI learns a new convention or preference during a conversation, it writes it down immediately, not at commit time. Nothing is lost if you commit from the command line without AI involvement.
Every file is queryable. A metadata standard on every document means your AI can scan summaries and tags without loading full files. This scales to hundreds of documents.
It maintains itself. Lint checks surface stale files, missing metadata, and orphaned docs. The system even reviews its own instruction efficiency to prevent context window bloat as your wiki grows.
What You Can Do
| Say this | What happens |
|---|---|
| "Capture this: here's how we set up monitoring..." | Creates a file in _inbox/, offers to file it |
| "We decided to use Terraform over Pulumi, document why" | Decision record with context, options, rationale, and an outcome section to fill in later |
| "What do we know about our hiring process?" | Scans summaries, synthesizes an answer with source citations |
| "Lint the wiki" | Surfaces stale files, missing metadata, orphaned docs |
| "Start a new initiative: migrate to Kubernetes" | Project workspace with brief, timeline, artifact tracking |
| "Create a slide deck summarizing Q1 decisions" | Generates output in _output/, dated and organized |
These form a cycle. Capture feeds queries, queries inform decisions, decisions get tracked, lint keeps it all healthy. The system compounds over time.
Go Deeper
| Topic | What's there |
|---|---|
| Training Phases | How the system adapts to you over 30 days |
| Directory Structure | What each folder is for, why there are only 4 |
| Journal Sync & Privacy | Syncing 3-Journal/ across devices — plaintext, git-crypt, or external |
| Metadata Standard | The format that makes everything queryable |
| Supported Tools | Claude Code, Cursor, Copilot, Windsurf, ChatGPT, and others |
| Obsidian Setup | Pre-configured vault, plugins, web clipper |
| Design Decisions | Why the system works the way it does |
| FAQ | Common questions |
| Linked Code Repos | Connect wiki project entries to real code repos — workspace files, context files, Obsidian plugin dev |
| Extending & Contributing | Customize locally vs. contribute upstream |
| Future Directions | Where this is headed |
| Philosophy | The deeper thinking behind the system |
The examples/ directory has four sample files showing the metadata standard in action. Read them, then delete them when you're comfortable.
Credits
Built on Andrej Karpathy's LLM Wiki Pattern. Implementation by Ariel Sakin, refined over several months of daily use managing business operations.
License
Apache 2.0. See LICENSE.
docs/journal-sync.md
Journal Sync and Privacy
3-Journal/ is where you write reflections, daily notes, and observations that aren't ready to share. By default, the template gitignores everything in 3-Journal/ except README.md, so journal entries live only on the machine where you wrote them. That's the safe default: new users don't accidentally push private reflections on the first commit. It also breaks the moment you use your wiki on more than one device. You'll write an entry on your laptop and not see it on your desktop.
Three options, from simplest to most secure. Pick one and commit to it. Don't mix them (see the warning at the bottom).
Option 1: Plaintext in a private repo
Remove the 3-Journal/* rule from .gitignore. Journal entries commit and sync like any other file in your repo. The content travels with the repo.
Safe only if your repo is, and stays, private forever. Git history is forever. Once you commit a journal entry, it sits in the repo's history. If the repo ever flips to public or leaks, every journal entry in history is exposed, including ones you later "deleted" (delete commits still sit in history behind the new HEAD).
Good fit: solo users with a private GitHub repo, comfortable betting on repo visibility never changing.
To switch: ask your AI to "implement option 1 from docs/journal-sync.md". The AI has enough spec in this doc to remove the gitignore rule, stage any existing journal files that were previously ignored, and commit.
Option 2: git-crypt with a passphrase-wrapped key
Use git-crypt to transparently encrypt journal files. Locally, the files look like regular markdown; in the repo, they're encrypted binary blobs. Your AI can still read and write them at session time (after unlock), git sees only ciphertext.
The usual git-crypt flow uses a binary key file that must be shipped securely between devices. A lighter variant: wrap that key with a passphrase (using age -p or gpg -c) and commit the wrapped blob to the repo. New device setup becomes "clone the repo, install git-crypt and age, decrypt the wrapped key with the passphrase, run git-crypt unlock."
Safer if the repo leaks, because the journal entries in history are encrypted blobs. An attacker with the repo still needs to crack the passphrase to read anything. Passphrase in your head is the single point of failure. Lose it, and every encrypted entry in history becomes unrecoverable. Binary merges can be fussy if you write the same journal file from two devices on the same day, so stick to one-file-per-day naming to minimize collisions.
Good fit: users who want cross-device sync but aren't comfortable betting on repo privacy staying intact.
To switch: ask your AI to "implement option 2 from docs/journal-sync.md". The AI has enough spec in this doc to install git-crypt and age if needed, generate the git-crypt key, wrap it with your passphrase, commit the wrapped blob, set up .gitattributes for 3-Journal/, and walk you through unlocking on a second device.
Option 3: Journal outside the repo
Leave 3-Journal/* gitignored. Put the real journal in a folder that's already synced by something else (iCloud, Dropbox, Syncthing, a private NAS) and symlink it into the repo's 3-Journal/ path.
No crypto, no git history for the journal itself, the cloud provider becomes the trust anchor. On the plus side: zero passphrase friction, no binary merges, sync conflicts handled by the sync tool instead of by git.
Good fit: users already paying for a trusted sync service who don't want the journal mixed into git at all.
To switch: ask your AI to "implement option 3 from docs/journal-sync.md". The AI has enough spec in this doc to help you pick a location, move any existing entries, set up the symlink, and verify the gitignore still catches the symlinked path so nothing leaks into git.
Don't partially migrate
Each "ask your AI" instruction above is self-contained: a fresh AI session reading this doc has enough spec to execute the chosen option without further design questions. But the options don't compose well.
If you start with Option 1 (plaintext) and later decide you want Option 2 (git-crypt), the plaintext entries stay in git history forever unless you rewrite history with git-filter-repo to strip them out before encrypting. That's a destructive operation on your repo and will break any clones. Similarly, moving from Option 1 to Option 3 leaves the plaintext in history even after you move the live files out.
Pick once, commit to it. If you're unsure, Option 3 is the least committing: it leaves the default gitignore rule alone, so switching later costs nothing in git history.
73 files
- .claude/CLAUDE.md
- .claude/settings.json
- .cursor/rules/bootstrap.mdc
- .cursorrules
- .github/copilot-instructions.md
- .gitignore
- .obsidian/app.json
- .obsidian/appearance.json
- .obsidian/core-plugins.json
- .obsidian/daily-notes.json
- .obsidian/graph.json
- .obsidian/snippets/llm-wiki.css
- .obsidian/templates.json
- .windsurfrules
- 1-Projects/README.md
- 1-Projects/_SHIPPED/.gitkeep
- 1-Projects/_parked/README.md
- 2-Knowledge/Decisions/.gitkeep
- 2-Knowledge/HowTo/.gitkeep
- 2-Knowledge/README.md
- 2-Knowledge/References/.gitkeep
- 2-Knowledge/log.md
- 3-Journal/README.md
- 4-Private/README.md
- ATTRIBUTION.md
- CODE_OF_CONDUCT.md
- CONTRIBUTING.md
- LICENSE
- PHILOSOPHY.md
- README.md
- WIKI-LOG.md
- _config/config.md
- _config/context.md
- _config/standard.md
- _config/tools.md
- _inbox/README.md
- _meta/instructions/agent-write.md
- _meta/instructions/blog-posts/instructions.md
- _meta/instructions/definition-of-done.md
- _meta/instructions/general.md
- _meta/instructions/knowledge-lint.md
- _meta/instructions/knowledge-query.md
- _meta/instructions/optimization-review.md
- _meta/templates/blog-post.md
- _meta/templates/decision.md
- _meta/templates/inbox-capture.md
- _meta/templates/initiative.md
- _meta/templates/journal-entry.md
- _meta/templates/knowledge-article.md
- _meta/templates/meeting-notes.md
- _output/README.md
- _sources/.gitkeep
- _sources/README.md
- assets/.gitkeep
- assets/clipper-template.json
- docs/claude-code-setup.md
- docs/design-decisions.md
- docs/directory-structure.md
- docs/extending-and-contributing.md
- docs/faq.md
- docs/future-directions.md
- docs/journal-sync.md
- docs/linked-code-repos.md
- docs/llm-wiki-pattern.md
- docs/metadata-standard.md
- docs/obsidian-setup.md
- docs/supported-tools.md
- docs/training-phases.md
- examples/example-decision-record.md
- examples/example-inbox-capture.md
- examples/example-initiative.md
- examples/example-knowledge-article.md
- migrate-to-yaml-frontmatter.py
Advantages
- The system adapts to user patterns over a 30-day training period without requiring upfront taxonomy design. evidence: README.md
- Metadata standards on every document allow the AI to scan summaries and tags without loading full files, enabling scalability. evidence: README.md
- The inbox-first capture system allows for frictionless entry of information with a 7-day TTL for later filing. evidence: README.md
Limitations
- The template approach disconnects the user from upstream updates unless the Framework Track setup is used. evidence: README.md
Security
Risk level: No risk found
This brain runs
Agent instructions that run or fetch something (1)
.claude/CLAUDE.md:**Exception:** If the user's first message is a `/init` command or a meta question about the repo setup, complete that first — then run the Session Start Protocol on the next exchange.
Attribution and licence
- First committed by
asakin
- Publisher
- Ariel Sakin (Community)
- Source
- asakin/llm-context-base · commit 3c096ca
- Licence
- Apache License 2.0 (files served here)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
- Contributors
- asakin, claude
AutoGineer is not affiliated with or endorsed by asakin.
Download
The ZIP holds the files AutoGineer serves, with ATTRIBUTION.md and the licence. Download llm-context-base.zip