LLM Knowledge Bases

LLM wiki kit

First committed by rvk7895

Published in rvk7895/llm-knowledge-bases. Publisher page

35 GitHub starsUpdated 2026-06-16

What it is

Claude Code plugin for building LLM-maintained Obsidian wikis from raw research — compile, query, lint, and evolve your personal knowledge base. Inspired by Karpathy's knowledge base workflow.

— the description by rvk7895

AutoGineer overview, from the brain’s own files

LLM Knowledge Bases is a plugin for Claude Code designed to automate the creation and maintenance of a personal knowledge base in Obsidian. By treating the LLM as a knowledge compiler, the system ingests raw materials—such as articles, papers, and transcripts—and organizes them into a structured, interlinked wiki. Users interact with the system by feeding raw data into a designated folder, while the agent handles the compilation, indexing, and health maintenance of the vault. It supports multiple query depths, ranging from quick index lookups to deep, multi-agent research pipelines, ensuring that the knowledge base evolves alongside the user's research interests.

How it works

Memory model: LLM wiki

AutoGineer overview, from the brain’s own files

The system operates by ingesting raw source documents from a raw folder. It then compiles these into an Obsidian vault containing summaries, backlinks, and indexes. Users can query the knowledge base at three levels of depth: quick, standard, or deep. The system also performs automated maintenance tasks like linting for broken links or orphans and provides tools for deep research, which involves generating outlines and deploying parallel agents to research specific topics. All outputs are saved to an output folder or integrated back into the wiki.

Folders
Folder	Role	Files
.claude	agent-config	1
docs	docs	2
.claude-plugin	folder	1
plugins	folder	12

Requirements

Works with: Claude Code, OpenCode

Set it up

This brain runs code automatically: 1 script. Review plugins/kb/setup.sh before opening the folder in Claude Code. No step below turns a hook on for you.
- plugins/kb/setup.sh: sh plugins/kb/setup.sh
Download the brain as a ZIP and unzip it where you keep your notes or project: https://mp-staging.autogineer.com/agent-brains/rvk7895/llm-knowledge-bases/download
Claude Code reads CLAUDE.md from this folder automatically.
cd llm-knowledge-bases && claude
AutoGineer scanned these files at commit 342686f.
Not scanned by AutoGineerThe author’s README also describes installing it as a Claude Code plugin marketplace. Read its install section: AutoGineer did not scan that path. https://github.com/rvk7895/llm-knowledge-bases/blob/342686f4cdf0e969872601eabbcc10466da91431/README.md#installation
First committed by rvk7895. Published in rvk7895/llm-knowledge-bases. Licence: MIT License.

Usage

AutoGineer overview, from the brain’s own files

Initialize a new knowledge base structure
Compile raw research documents into an Obsidian wiki
Query the knowledge base for information at varying depths
Perform automated health checks on the wiki
Execute deep research on specific topics using multi-agent pipelines

“/kb compile”: The agent processes all files in the raw/ directory and updates the Obsidian wiki with new summaries and backlinks.
“/research-deep Artificial Intelligence trends”: The agent generates a research outline and launches parallel agents to conduct deep research on each item in the outline.
“/kb lint”: The agent scans the wiki for broken links, orphans, and inconsistencies, reporting them for resolution.

Documentation by rvk7895

LLM Knowledge Bases

A Claude Code plugin that turns raw research material into an LLM-maintained Obsidian wiki -- inspired by Andrej Karpathy's description of using LLMs as knowledge compilers rather than just code manipulators.

You drop source material into raw/, run a single command, and Claude handles the rest: compiling interlinked wiki articles, maintaining indexes and backlinks, answering complex questions at multiple depth levels, and continuously improving the knowledge base over time.

The LLM owns the wiki. You rarely edit it manually -- just explore in Obsidian and keep feeding it raw data.

How It Works

Ingest -- Raw documents (articles, papers, repos, YouTube transcripts, images, datasets) go into raw/
Compile -- Claude builds a structured Obsidian vault with summaries, backlinks, concept articles, and auto-generated indexes
Query -- Three depth levels:
- Quick -- Answers from wiki indexes and summaries alone
- Standard -- Cross-references the full wiki, supplements with web search
- Deep -- Multi-agent research pipeline with parallel web search agents
Output -- Markdown reports, Marp slides, matplotlib charts -- saved to output/ and optionally filed back into the wiki
Maintain -- Automated health checks (broken links, orphans, inconsistencies) and suggestions for new articles and connections

Prerequisites

Claude Code (CLI), installed and authenticated
Obsidian

Obsidian plugins (Dataview, Obsidian Git, etc.) are installed automatically by the setup script during /kb-init.

Installation

In Claude Code, run:

/plugin marketplace add rvk7895/llm-knowledge-bases
/plugin install kb@llm-knowledge-bases

All skills are installed as a single plugin.

Quick Start

# Initialize a new knowledge base
/kb-init

# Add sources to raw/, then compile the wiki
/kb compile

# Query the knowledge base
/kb query

# Run a health check
/kb lint

# Deep multi-agent research on a topic
/research <topic>
/research-deep

Available Skills

Skill	Purpose
`kb-init`	One-time setup: scaffolds directories, generates config, writes project files
`kb`	Main skill with four workflows: compile, query, lint, evolve
`research`	Generate a structured research outline for a topic
`research-deep`	Launch parallel agents for deep research on each outline item
`research-add-fields`	Add field definitions to an existing research outline
`research-add-items`	Add research targets to an existing outline
`research-report`	Compile deep research results into a markdown report

Directory Structure (after `/kb-init`)

raw/           -- Raw source documents
wiki/          -- LLM-compiled Obsidian vault
output/        -- Query results, slides, charts, reports
kb.yaml        -- Configuration
CLAUDE.md      -- Project instructions for Claude

X/Twitter Integration (Optional)

To ingest tweets, threads, and bookmarks from X, install Smaug:

npm install -g @steipete/bird
git clone https://github.com/alexknowshtml/smaug && cd smaug && npm install
npx smaug setup  # configures X session cookies

Once configured, paste any x.com link and the kb skill will fetch and compile it automatically. Without Smaug, you can still add X content by pasting tweet text directly or using Thread Reader App for threads.

Note: Smaug uses X session cookies for read-only access to your own data. This is not officially supported by X's TOS. Practical risk for personal use is very low, but be aware.

Attribution

Andrej Karpathy -- Original vision for LLM-maintained knowledge bases
Weizhena -- Deep Research skills adapted for the research pipeline

plugins/kb/skills/research-report/SKILL.md

user-invocable: true description: Summarize deep research results into a markdown report, covering all fields, skipping uncertain values. allowed-tools: Read, Write, Edit, Glob, Bash, AskUserQuestion

Attribution: Originally authored by Weizhena. Included with attribution for use in the Deep query workflow.

Research Report - Summary Report

Trigger Method

/research-report

Execution Flow

Step 0: Load Report Preferences

Read the vault's kb.yaml and extract the report_preferences: block. These are free-text prose instructions written by the user at init (kb-init §5.5) or via /kb-preferences. They control audience, register, depth, code handling, diagrams, self-containment, citations, and argument iteration for this report.

If the block is present: apply every field as an instruction to the prose you write in subsequent steps. Treat the field text literally — it is the working rule, not a keyword list. The Python generator script written in Step 3 should also respect these preferences (e.g., if depth asks for multi-paragraph per-item walkthroughs, emit the item template with multi-paragraph slots; if diagrams says ASCII-only, don't emit mermaid).

If the block is missing: fall back to factory defaults in plugins/kb/references/report-style-guide.md silently and include this line in the final report output:

No report_preferences set in kb.yaml — using factory defaults. Run /kb-preferences init to customize.

If kb.yaml itself is missing: tell the user to run kb-init and stop.

Per-task overrides. If the user's current request explicitly contradicts a stored preference ("make this one short", "skip the diagrams for this report"), follow the request for this report only. Do NOT modify kb.yaml — that's what the reflection step is for.

Step 1: Locate Results Directory

Find */outline.yaml in current working directory, read topic and output_dir configuration.

Step 2: Scan Optional Summary Fields

Read all JSON results and extract fields suitable for display in the table of contents (numeric, short indicators), such as:

github_stars
google_scholar_cites
swe_bench_score
user_scale
valuation
release_date

Use AskUserQuestion to ask the user:

Which fields should be displayed in the table of contents besides item names?
Provide dynamic option list (based on fields actually present in JSONs)

Step 3: Generate Python Conversion Script

Generate generate_report.py in the {topic}/ directory with requirements:

Read all JSONs from output_dir
Read fields.yaml to get field structure
Cover all field values from each JSON
Skip fields containing [uncertain] values
Skip fields listed in uncertain array
Generate markdown report format: table of contents (with anchor jumps + user-selected summary fields) + detailed content (organized by field categories)
Save to {topic}/report.md

Table of Contents Format Requirements:

Must include every item
Each item displays: sequence number, name (anchor link), user-selected summary fields
Example: 1. [GitHub Copilot](#github-copilot) - Stars: 10k | Score: 85%

Script Technical Points (Must Follow)

1. JSON Structure Compatibility Support two JSON structures:

Flat structure: fields directly at top level {"name": "xxx", "release_date": "xxx"}
Nested structure: fields in category sub-dicts {"basic_info": {"name": "xxx"}, "technical_features": {...}}

Field lookup order: top level -> category mapping key -> traverse all nested dicts

2. Category Multilingual Mapping Category names in fields.yaml and JSON keys can be any combination (Chinese-Chinese, Chinese-English, English-Chinese, English-English). Must establish bidirectional mapping:

CATEGORY_MAPPING = {
    "Basic Info": ["basic_info", "基本信息"],
    "Technical Features": ["technical_features", "technical_characteristics", "技术特性"],
    "Performance Metrics": ["performance_metrics", "performance", "性能指标"],
    "Milestone Significance": ["milestone_significance", "milestones", "里程碑意义"],
    "Business Info": ["business_info", "commercial_info", "商业信息"],
    "Competition & Ecosystem": ["competition_ecosystem", "competition", "竞争与生态"],
    "History": ["history", "历史沿革"],
    "Market Positioning": ["market_positioning", "market", "市场定位"],
}

3. Complex Value Formatting

list of dicts (like key_events, funding_history): format each dict as one line, separate kv with |
Regular lists: short lists joined with commas, long lists displayed with line breaks
Nested dicts: recursively format, display with semicolons or line breaks
Long text strings (over 100 characters): add line breaks <br> or use blockquote format for better readability

4. Additional Field Collection Collect fields present in JSON but not defined in fields.yaml, put into "Other Information" category. Note filtering:

Internal fields: _source_file, uncertain
Nested structure top-level keys: basic_info, technical_features, etc.
uncertain array: display each field name line by line, don't compress into one line

5. Uncertain Value Skipping Skip conditions:

Field value contains [uncertain] string
Field name is in uncertain array
Field value is None or empty string

Step 4: Execute Script

Run python {topic}/generate_report.py

Step 5: Preference Reflection

After report.md is written, scan the conversation for preference signals — moments where the user corrected your approach, accepted a non-obvious style choice silently, or said "remember this" / "for every work I do, remember X".

Follow the Reflection Protocol defined in plugins/kb/skills/kb-preferences/SKILL.md:

Scan for signals tied to specific report_preferences fields.
If signals found, propose concrete edits (show old-value excerpt → new-value excerpt).
Present proposals as a numbered list. Ask the user: [a] all / [n] none / [e] edit each / [1,3] subset.
On approval, write updates to kb.yaml and append entries to .meta/_preference_history.md with trigger reflection after /research-report.
If no signals found, skip silently.

Signal strength rules (from kb-preferences):

Don't propose a change for a single offhand remark. Require explicit "remember this" phrasing, two occurrences, or feedback directly tied to this report.
Don't propose contradictory changes. If the current field already captures the feedback, skip.
Keep field text under ~400 words total. If a field would bloat past that, start a notes bullet instead.

Output

{topic}/generate_report.py - Conversion script
{topic}/report.md - Summary report (styled per report_preferences: in kb.yaml)
Optionally: updates to kb.yaml + one or more appended entries in .meta/_preference_history.md if the reflection step produced approved changes

21 files

Advantages

Automates the maintenance of interlinked wiki articles and indexes, reducing manual effort. evidence: README.md
Supports multi-agent research pipelines for deep, parallel investigation of topics. evidence: README.md

Limitations

Requires Claude Code CLI and Obsidian to function. evidence: README.md
The LLM owns the wiki, and manual editing of articles outside of the provided workflows is discouraged. evidence: CLAUDE.md

Security

Risk level: Medium risk

This brain runs

Install scripts (1)

plugins/kb/setup.sh: sh plugins/kb/setup.sh · uses the network

Agent instructions that run or fetch something (44)

plugins/kb/skills/kb-init/SKILL.md: After confirming the vault path, run the bundled setup script via Bash tool:
plugins/kb/skills/kb-init/SKILL.md: The script is idempotent — safe to re-run.
plugins/kb/skills/kb-init/SKILL.md: python -c "import pptx" 2>/dev/null && echo "python-pptx: available" || echo "python-pptx: not installed (needed for slide extraction)"
plugins/kb/skills/kb-init/SKILL.md: python -c "import pandas" 2>/dev/null && echo "pandas: available" || echo "pandas: not installed (needed for dataset analysis)"
plugins/kb/skills/kb-init/SKILL.md: When the user sets up Smaug (during init or later), save the install path here.
plugins/kb/skills/kb-preferences/SKILL.md: If missing, tell the user to run `kb-init` and stop.
plugins/kb/skills/kb-preferences/SKILL.md: | Invocation | Mode | Action | |---|---|---| | `/kb-preferences` (bare) | interactive | Show current prefs, ask what to update | | `/kb-preferences init` | init | Re-run the 5 init questions, regenerate all fields from style-guide + user answers | | `/kb-preferences update <field>` | up …[+30 chars]
plugins/kb/skills/kb-preferences/SKILL.md: re-ask matching question or accept free-text | | `/kb-preferences add-note "..."` | add-note | Append the quoted string as a new line/paragraph under `notes:` | | `/kb-preferences reflect` | reflect-now | Run the reflection step against the current conversation explicitly | | `/kb-pref …[+164 chars]
plugins/kb/skills/kb-preferences/SKILL.md: If missing, tell user to run `/kb-preferences init` first and stop.
plugins/kb/skills/kb-preferences/SKILL.md: Confirm loudly: "This will replace all 9 fields with factory defaults and re-run the 5 init questions.
plugins/kb/skills/kb/SKILL.md: If missing, tell the user to run `kb-init` and stop.
plugins/kb/skills/kb/SKILL.md: Run `/kb-preferences init` to customize."
plugins/kb/skills/kb/SKILL.md: After a compile or query finishes writing its output, run a **lightweight reflection step** — only propose preference updates for **high-confidence signals**:
plugins/kb/skills/kb/SKILL.md: The user can run `/kb-preferences reflect` manually if they want a forced pass.
plugins/kb/skills/kb/SKILL.md: Run lightweight reflection at the end.
plugins/kb/skills/kb/SKILL.md: **Detection heuristic:** Run this check for each file:
plugins/kb/skills/kb/SKILL.md: Please provide a transcript, or install: pip install openai-whisper"
plugins/kb/skills/kb/SKILL.md: Try: `yt-dlp --write-auto-sub --sub-lang en --skip-download --output "/tmp/%(id)s" "<url>"` to get subtitles · uses the network
plugins/kb/skills/kb/SKILL.md: Try extracting with python-pptx: `python -c "from pptx import Presentation;
plugins/kb/skills/kb/SKILL.md: For Parquet: `python -c "import pandas as pd;
plugins/kb/skills/kb/SKILL.md: Run lightweight reflection at the end (after the answer is delivered;
plugins/kb/skills/kb/SKILL.md: Run the appropriate pipeline.
plugins/kb/skills/kb/SKILL.md: Run the graph health script if available:
plugins/kb/skills/kb/SKILL.md: python plugins/kb/scripts/graph_health.py "<vault_path>" --json --fix-suggestions
plugins/kb/skills/kb/SKILL.md: Run the evaluation script if available:
plugins/kb/skills/kb/SKILL.md: python experiments/evaluate.py "<vault_path>" --json
plugins/kb/skills/kb/SKILL.md: If user approves, execute the improvements
plugins/kb/skills/kb/SKILL.md: Re-run evaluation to measure improvement
plugins/kb/skills/kb/SKILL.md: Run: `npx smaug fetch <tweet_id>` then `npx smaug process`
plugins/kb/skills/kb/SKILL.md: **Install Smaug** (recommended) — `git clone https://github.com/alexknowshtml/smaug && cd smaug && npm install`, then `npx smaug setup` to configure X session cookies (`auth_token` + `ct0` from browser DevTools → Cookies → x.com). · uses the network
plugins/kb/skills/kb/SKILL.md: After setup, **save the install path to `kb.yaml`** under `integrations.smaug.path`.
plugins/kb/skills/kb/SKILL.md: Try to fetch transcript: `yt-dlp --write-auto-sub --sub-lang en --skip-download --output "/tmp/%(id)s" "<url>" 2>&1` · uses the network
plugins/kb/skills/kb/SKILL.md: If yt-dlp not installed: try `pip install yt-dlp` then retry
plugins/kb/skills/kb/SKILL.md: Extract video metadata: `yt-dlp --dump-json "<url>" 2>/dev/null | python -c "import sys,json; · uses the network
plugins/kb/skills/kb/SKILL.md: On the next `/kb compile` run (for compile-time context searches)
plugins/kb/skills/kb/SKILL.md: Forgetting to dispatch the auto-evolve subagent after a query -- it must always run, even if it usually exits silently
plugins/kb/skills/research-add-items/SKILL.md: Execute simultaneously:
plugins/kb/skills/research-deep/SKILL.md: Execute in batches by batch_size (need user approval to proceed to next batch after completing one)
plugins/kb/skills/research-deep/SKILL.md: After completing JSON output, run validation script to ensure complete field coverage:
plugins/kb/skills/research-deep/SKILL.md: python ~/.claude/skills/research/validate_json.py -f {fields_path} -j {output_path}
plugins/kb/skills/research-deep/SKILL.md: After completing JSON output, run validation script to ensure complete field coverage:
plugins/kb/skills/research-deep/SKILL.md: python ~/.claude/skills/research/validate_json.py -f /home/weizhena/AIcoding/aicoding-history/fields.yaml -j /home/weizhena/AIcoding/aicoding-history/results/GitHub_Copilot.json
plugins/kb/skills/research-report/SKILL.md: **If `kb.yaml` itself is missing:** tell the user to run `kb-init` and stop.
plugins/kb/skills/research-report/SKILL.md: Run `python {topic}/generate_report.py`

Medium risk: runs-install-script-network-variable-host (plugins/kb/setup.sh)
Medium risk: network-variable-host (plugins/kb/setup.sh:81)
Medium risk: runs-instruction-instruction-network (plugins/kb/skills/kb/SKILL.md:206)
Medium risk: runs-instruction-instruction-network (plugins/kb/skills/kb/SKILL.md:571)
Medium risk: runs-instruction-instruction-network (plugins/kb/skills/kb/SKILL.md:596)
Medium risk: runs-instruction-instruction-network (plugins/kb/skills/kb/SKILL.md:600)
Low risk: network-instructions (.claude/settings.local.json:14)
Low risk: scripts-present (plugins/kb/setup.sh)
Low risk: network-instructions (plugins/kb/skills/kb/SKILL.md:202)
Low risk: autonomy-phrase (plugins/kb/skills/kb/SKILL.md:402)

Attribution and licence

First committed by

rvk7895

Publisher

Ritvik Aryan Kalra (Community)

Source

rvk7895/llm-knowledge-bases · commit 342686f

Licence

MIT License (files served here)

ATTRIBUTION.md (written by AutoGineer)

Synced

2026-11-12

Contributors

ritvik-sarvam, rvk7895

AutoGineer is not affiliated with or endorsed by rvk7895.

Download

The ZIP holds the files AutoGineer serves, with ATTRIBUTION.md and the licence. Download llm-knowledge-bases.zip

What it is

How it works

Requirements

Set it up

Usage

Documentation by rvk7895

LLM Knowledge Bases

How It Works

Prerequisites

Installation

Quick Start

Available Skills

Directory Structure (after /kb-init)

X/Twitter Integration (Optional)

Attribution

user-invocable: true description: Summarize deep research results into a markdown report, covering all fields, skipping uncertain values. allowed-tools: Read, Write, Edit, Glob, Bash, AskUserQuestion

Research Report - Summary Report

Trigger Method

Execution Flow

Step 0: Load Report Preferences

Step 1: Locate Results Directory

Step 2: Scan Optional Summary Fields

Step 3: Generate Python Conversion Script

Script Technical Points (Must Follow)

Step 4: Execute Script

Step 5: Preference Reflection

Output

Advantages

Limitations

Security

This brain runs

Attribution and licence

Download

Directory Structure (after `/kb-init`)