Prompt Kit

Open Brain Extensions: Companion Prompts

Your Open Brain is running. Your agent reads from it, writes to it, remembers what you said last Tuesday. Now it's time to give it hands. These six prompts cover the full arc: figure out which extensions match your life, learn to navigate the OB1 repo where everything lives, populate your first tables with real data, audit whether your setup has both doors open, design your own extensions using the four principles from the article, and — if you build something worth sharing — package it as a contribution to the community.

Everything lives here: OB1 on GitHub

Note: If you're brand new to Open Brain, start with the original companion prompts first — they cover initial setup, memory migration, and building the capture habit. This kit assumes your Open Brain is already running and you're ready to extend it.

What's included:

  • Prompt 0: Star OB1 — Your first GitHub skill (takes 10 seconds, makes the repo easier to find for everyone)
  • Prompt 1: Extension Matchmaker — Interviews you about your actual life, recommends which extensions to build first, and points you to the exact GitHub page for each one
  • Prompt 2: GitHub Navigator — Teaches you how to read, navigate, and use the OB1 repo even if you've never touched GitHub before
  • Prompt 3: Extension Launcher — Walks you through populating your first extension table with real data from your life — adapts to whichever extension you picked
  • Prompt 4: Two-Door Audit — Evaluates your current Open Brain setup against the two-door principle and identifies where you're only using half the system
  • Prompt 5: Design Your Own Extension — Uses the four principles from the article (time-bridging, cross-category reasoning, proactive surfacing, judgment line) to help you architect custom extensions for problems the six built-in ones don't cover
  • Prompt 6: OB1 Contribution Builder — Turns something you built on your Open Brain into a properly structured community contribution that passes the repo's automated review

Tools: All prompts work with any MCP-connected AI (Claude, ChatGPT, Gemini, Grok). Prompt 3 requires your Open Brain MCP server to be connected. Prompts 2 and 6 benefit from an AI with web browsing capability for navigating the GitHub repo directly.

Prompt 0: Star OB1 on GitHub

Job: Your first GitHub skill. Takes 10 seconds. Makes OB1 easier for other people to find — and bookmarks it for you so you can get back to it anytime.

When to use: Right now. Before you do anything else.

What you'll get: A starred repo, a warm feeling, and a permanent bookmark in your GitHub account.

Here's the whole thing:

  1. Open github.com/NateBJones-Projects/OB1
  2. If you don't have a GitHub account, create one (free, takes 30 seconds)
  3. Click the ☆ Star button in the top right corner of the repo page. It's right next to "Fork" and "Watch"
  4. That's it. The star turns yellow. You're done.

Why this matters: GitHub stars are how open-source projects get discovered. Every star makes it more likely that someone else searching for "personal AI database" or "MCP extensions" finds OB1 instead of some SaaS platform that charges $20/month for something you built in 45 minutes. Stars also tell us how many people are actually using this — which helps us prioritize what to build next.

Bonus: Your starred repos show up under your GitHub profile → Stars tab. It's a permanent bookmark. Next time you need to find the repo, you don't have to search — it's already saved.

Now go do the prompts that actually require an AI.

Prompt 1: Extension Matchmaker

Job: Interviews you about your life situation — kids, house, job hunt, relationships, cooking — and generates a personalized build order from the six OB1 extensions, pointing you to the exact GitHub README for each recommendation.

When to use: Right after reading the article, when you're staring at six use cases and not sure which one to start with. Also useful if your life situation changes (new job hunt, new house, kids starting school).

What you'll get: A ranked list of which extensions matter most for YOUR life, a recommended build order, and direct links to each extension's guide in the OB1 repo.

Output feeds into: Prompt 2 (GitHub Navigator) if you're new to GitHub, or Prompt 3 (Extension Launcher) if you're ready to build.

What the AI will ask you:

  1. Your living situation (homeowner/renter, kids, partner)
  2. Your professional situation (employed, job hunting, freelance, managing relationships)
  3. What annoys you most about managing your life right now
  4. Which use cases from the article hit hardest

Prompt 2: GitHub Navigator

Job: Teaches you how to read, navigate, and use the OB1 repository on GitHub — even if you've never used GitHub before. Walks you through the repo structure, explains what each folder and file does, and shows you how to find exactly what you need.

When to use: Before building your first extension, if GitHub feels unfamiliar. Also useful anytime you're looking for something specific in the repo and can't find it.

What you'll get: A guided tour of the OB1 repo structure, an understanding of how to read README files and find SQL schemas, and confidence to navigate the repo on your own.

Output feeds into: Prompt 3 (Extension Launcher) — once you can find the build guide for your chosen extension, you're ready to populate it with data.

What the AI will ask you:

  1. Your comfort level with GitHub (never used it / used it a little / comfortable)
  2. What you're looking for in the repo (specific extension, general orientation, or something else)

OB1/ ├── README.md ← Start here. Overview of everything. ├── CONTRIBUTING.md ← Rules for contributing back to the repo ├── extensions/ ← The six extension build guides │ ├── household-knowledge/ │ ├── home-maintenance/ │ ├── family-calendar/ │ ├── meal-planning/ │ ├── professional-crm/ │ └── job-hunt-pipeline/ ├── primitives/ ← Shared building blocks (RLS, shared MCP access) ├── setup/ ← Initial setup if you haven't built Open Brain yet ├── companion-prompts/ ← AI prompts that help you use extensions ├── recipes/ ← Community-contributed recipes ├── schemas/ ← Community-contributed table schemas ├── dashboards/ ← Community-contributed dashboard templates └── integrations/ ← Community-contributed integrations

Prompt 3: Extension Launcher

Job: Walks you through populating your first extension table with real data from your actual life. Adapts to whichever extension you chose — household knowledge, home maintenance, family calendar, meal planning, professional CRM, or job hunt pipeline.

When to use: After you've created the table from the OB1 build guide and you're staring at an empty schema. This prompt fills it with the first 10-20 entries by interviewing you about your real situation.

What you'll get: Your extension table populated with real, structured data — entries your agent can immediately start reasoning about.

Output feeds into: N/A — your data is now live in your Open Brain. Start asking your agent questions about it.

What the AI will ask you:

  1. Which extension you're populating
  2. Confirmation that the table exists in your Open Brain
  3. A structured interview specific to that extension (appliances and warranties for home maintenance, contacts and relationships for CRM, etc.)

Prompt 4: Two-Door Audit

Job: Evaluates your current Open Brain setup against the two-door principle from the article — agent door and human door — and identifies where you're only using half the system.

When to use: After you've been using Open Brain for a week or more and you want to know where the gaps are. Also useful after building a new extension to make sure both doors are working.

What you'll get: A diagnostic of which of the four modes you're actually using (agent reads, agent writes, human reads, human writes) across your Open Brain setup, with specific recommendations for closing the gaps.

Output feeds into: N/A — standalone diagnostic. Recommendations may point you to specific extensions or to Prompt 5 (Design Your Own Extension).

What the AI will ask you:

  1. What Open Brain extensions or tables you've built so far
  2. How you're currently interacting with your data (which AI clients, any visual interfaces, mobile access)
  3. Where the system feels useful vs. where it feels like it's not earning its keep

Prompt 5: Design Your Own Extension

Job: Helps you architect a custom Open Brain extension for a problem the six built-in ones don't cover, using the four principles from the article: time-bridging, cross-category reasoning, proactive surfacing, and the judgment line.

When to use: When you have a recurring problem in your life or work that involves scattered information, events spread over time, or multiple categories that need cross-referencing — and none of the existing extensions fit.

What you'll get: A complete extension design: table schema, data capture plan, example queries your agent should handle, and a clear judgment line for what the agent surfaces vs. what you decide.

Output feeds into: Prompt 6 (OB1 Contribution Builder) if you build something good enough to share.

What the AI will ask you:

  1. The problem you're trying to solve (what keeps falling through the cracks)
  2. Where the relevant information currently lives (scattered across which tools, people, memories)
  3. What decisions this data should inform
  4. Who else needs to see or interact with this data

Prompt 6: OB1 Contribution Builder

Job: Helps you turn something you built on your Open Brain into a properly structured community contribution to the OB1 repo — formatted correctly, documented clearly, and ready to pass the automated review.

When to use: When you've built a custom extension, recipe, schema, dashboard template, or integration that worked well for you and you think others could use it.

What you'll get: A complete contribution package: the right file structure, a README that matches OB1 standards, and guidance on submitting a pull request.

Output feeds into: N/A — your contribution goes to the OB1 repo.

What the AI will ask you:

  1. What you built and which contribution category it fits (recipe, schema, dashboard, integration)
  2. Your technical comfort with GitHub (the PR submission process varies by skill level)
  3. The details of what you built — SQL, configuration, workflow

Automated review rules the contribution must pass:

  1. README.md exists and is non-empty
  2. README follows the standard template structure
  3. SQL files use standard PostgreSQL syntax
  4. No hardcoded personal data (API keys, names, emails)
  5. File names use lowercase-with-hyphens convention
  6. Folder is in the correct contribution category directory
  7. No duplicate of existing contribution (by name or function)
  8. All referenced files exist in the folder
  9. Code blocks are properly formatted
  10. Prerequisites section lists dependencies
  11. Difficulty level is specified

Review each rule against the contribution and fix any issues before finalizing. </packaging>

<submission-guide> Based on the user's GitHub comfort level:

For GitHub beginners (never submitted a PR):

Walk them through these steps:

  1. Create a GitHub account if they don't have one
  2. Navigate to the OB1 repo: https://github.com/NateBJones-Projects/OB1
  3. Fork the repo (explain what this means: "Creating your own copy that you can edit")
  4. Create the folder and files in their fork using GitHub's web interface (click "Add file" → "Create new file")
  5. Write a clear commit message
  6. Submit a pull request from their fork to the main repo
  7. Explain what happens next: automated review runs, then a human admin reviews

For GitHub users:

Focus on:

  • Correct folder placement and naming
  • PR title convention
  • Making the contribution self-contained (no external dependencies that aren't documented) </submission-guide>
<guardrails> - Strip all personal data from the contribution. Names, emails, API keys, personal details — none of this goes in the repo. Replace with generic examples. - The README must be clear enough that someone with zero context about the contributor's setup can follow it. - SQL must be standard PostgreSQL that works in Supabase. No proprietary extensions or functions that aren't available in Supabase. - If the contribution overlaps significantly with an existing extension or community contribution, tell the user — they might be better off improving the existing one rather than adding a near-duplicate. - Don't submit the PR for them. Walk them through it, review everything, but they click the buttons. - If the contribution is too raw or incomplete to pass review, be honest: "This isn't ready yet. Here's what needs to happen before it can be submitted: [specific list]." </guardrails> ```

Why These Prompts Exist

Your Open Brain was infrastructure. These extensions make it operational. The Matchmaker tells you where to start. The Navigator gets you comfortable in the repo where everything lives. The Launcher fills your first table with real data. The Audit catches the gaps in your setup. The Designer lets you go beyond the built-in six. And the Contribution Builder means the best things people create come back to the community so everyone's Open Brain gets better.

Pick the use case that hit you hardest. Start there. The repo is at github.com/NateBJones-Projects/OB1 — and if this is your first time on GitHub, Prompt 2 has your back.

Built by Nate B. Jones — companion to "Your Open Brain Can Think. Now It Needs Hands."