# AI use cases for Second Brain OS (undefined-ui/second-brain-os) Repo: [undefined-ui/second-brain-os](https://github.com/undefined-ui/second-brain-os) | Use cases: 4 | Page: https://dedupedai.com/use-cases/undefined-ui-second-brain-os/ A starter kit and guide for building a knowledge base in plain markdown files that an AI agent builds, links and maintains for you, using Claude Code and Obsidian. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Turn saved articles into a self-linking wiki: https://dedupedai.com/use-cases/undefined-ui-second-brain-os/build-self-linking-wiki-from-saved-articles/ - Run a project through an Inputs, Process, Outputs pipeline: https://dedupedai.com/use-cases/undefined-ui-second-brain-os/run-one-project-through-inputs-process-outputs/ - Audit an agent setup with the agents course plugin: https://dedupedai.com/use-cases/undefined-ui-second-brain-os/audit-agent-context-with-course-plugin/ - Feed live calendar, email and chat into the vault: https://dedupedai.com/use-cases/undefined-ui-second-brain-os/connect-live-calendar-email-chat-to-vault/ ## How to turn saved articles into a self-linking wiki with Second Brain OS Repo: [undefined-ui/second-brain-os](https://github.com/undefined-ui/second-brain-os) | Difficulty: Intermediate | Time: 60 minutes | Basis: From the README Copy the starter vault, clip an article, and let the agent file it into linked wiki pages. **Who it is for:** A researcher or engineer who saves articles and videos all week and never rereads them, and wants them filed and linked without manual work. **The problem:** Bookmarks and read-later queues pile up unconnected. Filing and linking them by hand stops after a few weeks. **The outcome:** A local vault of markdown files where each clipped source becomes linked wiki pages that update existing pages instead of piling up beside them. ### Steps 1. `git clone https://github.com/undefined-ui/second-brain-os.git` 2. `cp -r second-brain-os/vault-template ~/brain` 3. `mkdir -p ~/brain/.claude` then `cp -r second-brain-os/skills ~/brain/.claude/skills`, `cp -r second-brain-os/commands ~/brain/.claude/commands`, `cp -r second-brain-os/agents ~/brain/.claude/agents` and `cp -r second-brain-os/scripts ~/brain/scripts` (inferred, check the README) 4. `rm ~/brain/.claude/*/README.md ~/brain/scripts/README.md` 5. `cd ~/brain && claude`, then clip one article into `raw/` with the Web Clipper and run `/ingest` 6. Open `~/brain` in Obsidian with "Open folder as vault" and check the new pages and their links ### Prompt to paste into your agent ```text Using the second-brain-os repo, copy vault-template to ~/brain and copy skills, commands, agents and scripts into ~/brain/.claude and ~/brain/scripts. Place one article in ~/brain/raw/, run /ingest, and report the wiki pages created. Verify that every new page has at least one wikilink to another page. ``` ### Caveat The README says the vault is built around Claude Code and Obsidian, so other agents will need the same markdown layout, and the folder READMEs must be removed so the agent does not read them as vault content. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/undefined-ui-second-brain-os/build-self-linking-wiki-from-saved-articles/ ## How to run a project through an Inputs, Process, Outputs pipeline with Second Brain OS Repo: [undefined-ui/second-brain-os](https://github.com/undefined-ui/second-brain-os) | Difficulty: Intermediate | Time: 60 minutes | Basis: From the README Give one project its own CLAUDE.md and a four-stage folder pipeline the agent follows. **Who it is for:** A writer, engineer or student who is about to ship one piece of work and wants the agent to use the vault's notes while doing it. **The problem:** Project notes live apart from what the agent knows, so each session starts from zero and the work drifts from the source material. **The outcome:** A scoped project folder with its own CLAUDE.md and Inputs, Process, Outputs and Feedback stages that draws on the wiki and feeds results back into it. ### Steps 1. Complete the vault setup from the first use case so `~/brain` exists with `.claude` and `scripts` in place 2. Create one project folder inside `~/brain` for the work you are shipping, following the project pipeline in `vault-template` (inferred, check the README) 3. Add a `CLAUDE.md` in that project folder that states its goal and points to the relevant wiki pages (inferred, check the README) 4. `cd ~/brain && claude` and work through the project's Inputs, Process, Outputs and Feedback stages in order 5. Record the outcome in the Feedback stage so the next session can read it ### Prompt to paste into your agent ```text In the second-brain-os vault at ~/brain, work inside the project using its CLAUDE.md and its Inputs, Process, Outputs and Feedback folders. Produce the project's output file and note what changed in Feedback. Verify that the output cites at least two existing wiki pages. ``` ### Caveat The README describes the project pipeline in general terms, so the exact folder names and file contents should be taken from the vault-template files rather than assumed. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/undefined-ui-second-brain-os/run-one-project-through-inputs-process-outputs/ ## How to audit an agent setup with the agents course plugin with Second Brain OS Repo: [undefined-ui/second-brain-os](https://github.com/undefined-ui/second-brain-os) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Install the course's Claude Code plugin in two commands and run its context auditor. **Who it is for:** A developer building a Claude Code agent who suspects the context it loads is bloated or badly ordered and wants a checklist-based review. **The problem:** Agent context grows without anyone checking it, and slow or confused agents are hard to trace back to what they were given. **The outcome:** The course plugin installed in Claude Code, with its context auditor run against your repo to produce a review of what the agent sees. ### Steps 1. `claude plugin marketplace add undefined-ui/second-brain-os` 2. `claude plugin install agents-course@second-brain-os` 3. Open the repo you want reviewed in Claude Code and ask for the context auditor by name (inferred, check the README) 4. Read the auditor's findings against Module 1 of the agents course before changing anything ### Prompt to paste into your agent ```text Install the agents-course plugin from undefined-ui/second-brain-os in Claude Code, then run its context auditor on . Output the auditor's findings as a list of context problems with file paths. Verify that each finding names a specific file. ``` ### Caveat The README gives the plugin one line per tool, so the exact invocation of the auditor is not stated and must be confirmed from the plugin once it is installed. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/undefined-ui-second-brain-os/audit-agent-context-with-course-plugin/ ## How to feed live calendar, email and chat into the vault with Second Brain OS Repo: [undefined-ui/second-brain-os](https://github.com/undefined-ui/second-brain-os) | Difficulty: Advanced | Time: 90 minutes | Basis: From the README Follow the live-data guide to let the agent read your calendar, email and chat. **Who it is for:** A knowledge worker who wants the vault to know what is on their schedule and in their inbox, not only what they clipped. **The problem:** The vault only holds what you saved by hand, so it misses the meetings and messages that shape the work you are doing. **The outcome:** A vault that can read live calendar, email and chat sources through the setup described in the live-data guide. ### Steps 1. Complete the vault setup from the first use case so `~/brain` is running in Claude Code 2. Open the live-data setup guide at `docs/02-setup/live-data.md` in the repo 3. Connect the calendar, email and chat sources that guide describes, one at a time 4. Run `cd ~/brain && claude` and confirm the agent can read one item from each source ### Prompt to paste into your agent ```text Using the live-data guide in the second-brain-os repo, connect the calendar, email and chat sources for the vault at ~/brain. Report which sources connected and list one item read from each. Verify that no credential is written into any markdown file in the vault. ``` ### Caveat The README only links to a live-data guide for these sources, so the specific connection steps must be taken from that page rather than from this summary. Risk, credentials: Calendar, email and chat access needs account credentials, so keep them out of the vault and out of any file the agent writes into it. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/undefined-ui-second-brain-os/connect-live-calendar-email-chat-to-vault/