Second Brain skill
Four skills for a personal Obsidian knowledge base: an agent compiles articles and notes into a linked wiki, answers from it and health-checks it
- 764 ★ on the repository
- Last commit
Extended access
Why this level
- The agent creates and edits many Markdown files in the base folder and writes config files for the chosen agents
- The wizard offers to install third-party tools and runs an onboarding script
- Your sources and notes are read by the agent's model and its provider
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/nicholasspisak-second-brainDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The agent creates and edits many Markdown files in the base folder and writes config files for the chosen agents
- The wizard offers to install third-party tools and runs an onboarding script
- Your sources and notes are read by the agent's model and its provider
This is third-party code. Review the repository files before installing.
What it does
Second Brain builds a personal knowledge base in Obsidian on the pattern of a wiki maintained by a language model. You drop raw material (articles, Web Clipper pages, notes, transcripts) into a raw folder, the agent reads it and writes structured pages with cross-references, and you browse the result in Obsidian. The `/second-brain` skill runs a five-step setup wizard: base name (second-brain by default), location (~/Documents/ by default), topic, config files for the agents you use and extra tools. It creates the raw, wiki and output folders and an agent config file such as CLAUDE.md. The `/second-brain-ingest` skill processes new files from raw. The agent reads a source in full, states 3–5 key takeaways and waits for your confirmation, then writes a summary page in wiki/sources, updates or creates pages for people, organizations, tools and concepts, adds wiki-style [[links]] and appends to the index and operation log. It flags contradictions with existing content, citing both sources. The `/second-brain-query` skill answers questions from the wiki: it reads the index first, uses the qmd search for large bases, cites pages with links and offers to save a valuable answer as a synthesis page. The `/second-brain-lint` skill checks the wiki for broken links, orphan pages, contradictions, stale claims, missing pages and index mismatches.
Who it is for. For researchers, analysts, writers and anyone who collects articles and notes in Obsidian and wants an agent to maintain a linked knowledge base from them.
Good fit when
- A folder of articles and notes has piled up and you need a linked wiki with separate pages for concepts and people
- You need to answer a question from your own materials only, with links to wiki pages
- The wiki has grown and you need to find broken links, pages with no incoming links and contradictions between sources
- One set of notes should be maintained by several agents, such as Claude Code and Codex, under a shared schema
- You want comparisons and analyses from answers saved back into the base as synthesis pages
Not a fit when
- You do not use Obsidian: the folder structure and [[wiki links]] are built around it
- You need fully automatic ingestion without your involvement: before writing pages the agent discusses key takeaways and waits for confirmation
- You need shared server-side search for a team: the base is a plain folder on your computer
- You have few sources: the scheme pays off when enough pages and links accumulate for search and checks
Example request
I put three articles about vector databases into raw. Ingest them into the wiki, then tell me where their approaches differLimitations
The author did not specify a license, so rights to use and copy are undefined; check with the author.
It needs Obsidian, an agent that supports skills and Node.js for installing through npx. The wizard installs the optional tools summarize, qmd and agent-browser only with your consent.
In Obsidian settings, clipped attachments must be pointed at raw/assets/ by hand, and images downloaded with a command in the app itself.
The wiki is not rebuilt from raw automatically: if a file in raw changes after ingestion, the pages stay built from the original content. The author advises running `/second-brain-lint` after every ten ingests or monthly. For large wikis qmd speeds up queries, installed with `npm i -g @tobilu/qmd`.
How to disable. Remove the four second-brain skills from your agent's skills folder or with the same manager you installed with. The knowledge base itself (the raw, wiki and output folders) does not need deleting: these are plain Markdown files that stay in Obsidian. The agent config file inside the base folder can be removed by hand.
FAQ
What is in the knowledge base folders?
The raw folder is the inbox: sources go there and attachments go to raw/assets. The wiki folder holds source summaries, pages for people, organizations, concepts and syntheses, plus index.md and log.md. Reports and other results go to output.
How do I rerun the wizard if it failed?
Call `/second-brain` again: the onboarding script is idempotent and does not overwrite existing files, so your data stays. For a fresh start delete the base folder and run the wizard anew.
What if I accidentally changed a file in raw?
The wiki was built from the original content, so its pages are unaffected. You can restore the original from git history if the base lives in a repository, or clip the source again.
How do I speed up queries on a large wiki?
Install qmd with `npm i -g @tobilu/qmd`. The query skill uses it automatically when present and searches the wiki files with hybrid search.
Can several agents maintain one base?
Yes, the wizard creates config files for the agents you pick (Claude Code, Codex, Cursor, Gemini CLI), and all of them work under the same wiki schema.
How often should I lint the wiki?
The author advises after every ten ingests or monthly, and before a big query or synthesis. The skill groups findings by severity: errors, warnings and suggestions.
Discussion
No questions or reviews yet.
Related
SkillsAn Obsidian vault template with 18 skills, 72 commands and 6 subagents for Claude Code: the agent turns sources into linked pages and answers from them
A skill set that lets an agent run a personal Obsidian wiki: it compiles knowledge from documents and past sessions and answers with citations
Persistent agent memory in an Obsidian vault: saving decisions, search, research and background vault upkeep
A plugin and skill set that turns an Obsidian vault into a knowledge base with sources, links and grounded answers
A set of skills and agents for a second brain on markdown and git: notes, a knowledge base and a people CRM with no database or cloud
Obsidian skills
Obsidian Skills
Skills for agents working with Obsidian: Markdown with wikilinks, Bases, JSON Canvas, Obsidian CLI, Defuddle and Knap
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| Second Brain (this entry) | no | 764 | |
| LLM Wiki | no | 1,479 | plugin for Claude Code and Codex with topics, import and the /wiki:ingest and /wiki:archive commands |
| LLM Wiki Agent | no | 3,663 | workflow where the agent adds sources to a persistent linked wiki on its own |
README in short
The description lays out the idea of a wiki maintained by a language model after Andrej Karpathy's note: the model is the librarian and you are the curator. It then covers prerequisites, one-command installation, a table of the four skills, a seven-step quickstart, the base folder layout and optional tools. At the end it answers common questions on a failed wizard, edits to raw, index drift, broken links, large wikis, multiple agents and images in clips. The repository also holds the requirements document the project grew from.
README badge
Original SKILL.md
---
name: second-brain
description: >
Set up a new Obsidian knowledge base with the LLM Wiki pattern. Use when
the user wants to create a second brain, initialize a vault, set up a
personal knowledge base, or says "onboard". Guides through an interactive
wizard to configure vault name, location, domain, agent support, and tooling.
allowed-tools: Bash Read Write Glob Grep
---
# Second Brain — Onboarding Wizard
Set up a new Obsidian knowledge base using the LLM Wiki pattern. The LLM acts as librarian — reading raw sources, compiling them into a structured interlinked wiki, and maintaining it over time.
## Wizard Flow
Guide the user through these 5 steps. Ask ONE question at a time. Each step has a sensible default — the user can accept it or provide their own value.
### Step 1: Vault Name
Ask:
> "What would you like to name your knowledge base? This will be the folder name."
> Default: `second-brain`
Accept any user-provided name. This becomes the folder name and the title in the agent config.
### Step 2: Vault Location
Ask:
> "Where should I create it? Give me a path, or I'll use the default."
> Default: `~/Documents/`
Accept any absolute or relative path. Resolve `~` to the user's home directory. The final vault path is `{location}/{vault-name}/`.
### Step 3: Domain / Topic
Ask:
> "What's this knowledge base about? This helps me set up relevant tags and describe the vault's purpose."
>
> Examples: "AI research", "competitive intelligence on fintech startups", "personal health and fitness"