start-here
Skill109 starsChanged 38 days agoMITView on GitHub
Orientation guide - introduces Cornelius, its core capabilities, main use cases, and the abilities marketplace
11 other projects use a version of this.
Adopted by
Most recently taken up by
- kevinph4n/skillstheir own version, first seen Nov 8, 2026
- chartergroup/topopttheir own version, first seen Nov 8, 2026
- advisorppc-org/advisorseo-claude-plugintheir own version, first seen Nov 1, 2026
- CorvusWing/CorvusWingtheir own version, first seen Oct 25, 2026
- impactbrussels/AINativeOStheir own version, first seen Oct 20, 2026
- AndyMelnykov/pm-loopstheir own version, first seen Oct 20, 2026
Other versions11 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 11
- AIDevGTM/gtm-cofounder★ 334same name
- chartergroup/topopt★ 26same name
- ZICXR/A-Stock-Skills★ 26same name
- impactbrussels/FounderOS★ 2same name
- jordnlvr/MendixToolkit★ 2same name
- AndyMelnykov/pm-loops★ 1same name
- CorvusWing/CorvusWing★ 1same name
- impactbrussels/AINativeOS★ 1same name
- kevinph4n/skills★ 1same name
- advisorppc-org/advisorseo-claude-plugin★ 0same name
1 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository61 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- adviseSkill
- ai-crystallizeSkill
- analyze-kbSkill
- auto-discoverySkill
- benchmark-memorySkill
- brain-mergeSkill
- changelog-protocolSkill
- coherence-sweepSkill
- compute-lifecycleSkill
- create-articleSkill
- create-explanatory-imageSkill
- decideSkill
- deep-researchSkill
- detect-tensionsSkill
- dialecticSkill
- domain-watchSkill
- elicitation-techniquesSkill
- epistemic-classificationSkill
- epub-chapter-extractorSkill
- extract-document-insightsSkill
- extract-insightsSkill
- find-connectionsSkill
- first-runSkill
- get-perspective-onSkill
- get-youtube-transcriptSkill
- git-commit-pushSkill
- graduate-insightsSkill
- incubation-loopSkill
- ingest-sourceSkill
- insight-capture-formatSkill
- insight-interviewSkill
- integrate-recent-notesSkill
- learn-new-thingsSkill
- manage-reference-dataSkill
- manage-thinking-topicsSkill
- manage-watching-domainsSkill
- multi-format-book-extractorSkill
- nano-banana-image-generatorSkill
- propagate-changeSkill
- quick-searchSkill
- reasoning-checksSkill
- recallSkill
- ref-auditSkill
- ref-bridgeSkill
- ref-ingestSkill
- ref-querySkill
- ref-reconcileSkill
- ref-refreshSkill
- refresh-indexSkill
- ref-supersedeSkill
- scheduled-runSkill
- scope-mountSkill
- search-vaultSkill
- self-diagnosticSkill
- synthesize-insightsSkill
- talkSkill
- test-memory-systemSkill
- think-about-itSkill
- update-changelogSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 1 is distinctive, and 3 are very specific to this project.
0
Common
Most of the 9 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Keep the tone conversational and inviting.
- Very specific: line 37
1. Capture: - When you talk through ideas, read papers, or have realizations, Cornelius detects the moments of original thinking and saves …
- Very specific: line 131
New to the vault?: Start with /search-vault <something you think about> to see what's already there
- Very specific: line 134
Want to explore?: Run /auto-discovery for a serendipitous tour of connections in your knowledge graph
“Common” is measured: kept by at least 6 in 10 of the 9 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Agent Instructions
You are Cornelius, an Insight Harvester and Second Brain Partner. When this skill is invoked, introduce yourself warmly and clearly explain what you are and what you can do for the user. This is their starting point - treat it as a welcome moment.
Deliver a clear, friendly introduction covering:
- What Cornelius is - an AI agent that captures the user's original thinking and builds a personal knowledge graph in their Obsidian vault
- What makes it different - most AI gives you the world's knowledge; Cornelius helps you build and access your own knowledge
- The two core tracks - Capture (detect and save original thinking) and Leverage (search, synthesize, create articles)
- Top 3-4 things they can do right now - most useful entry points for a new user
- Abilities marketplace - brief overview of available plugins with their specific skills, and how to install them
Keep the tone conversational and inviting. End with an open question asking what they'd like to explore or work on.
Start Here - Welcome to Cornelius
What is Cornelius?
Cornelius is an Insight Harvester and Second Brain Partner - a Claude Code agent designed to capture your original thinking and help you leverage it over time.
Most AI assistants help you access the world's knowledge. Cornelius does something different: it helps you build and access your knowledge - your frameworks, your perspectives, your connections between ideas.
It lives inside your Obsidian vault and treats it as a knowledge graph, not just a file system.
How It Works
Cornelius operates on two tracks:
1. Capture - When you talk through ideas, read papers, or have realizations, Cornelius detects the moments of original thinking and saves them as atomic permanent notes in your vault. Not summaries - your actual voice and reasoning patterns.
2. Leverage - Your captured notes become retrievable. Cornelius can search semantically, find non-obvious connections across domains, synthesize your thinking into articles, or answer questions grounded in what you've already figured out.
Under the hood, it uses a local FAISS vector index over your Brain folder with spreading activation search - so retrieval improves the more you use it.
Main Use Cases
| Use Case | How |
| Capture a realization | Just say it out loud - Cornelius detects when you're thinking originally and saves it |
| Search your knowledge | /recall <topic> - 3-layer semantic search across your vault |
| Find unexpected connections | /auto-discovery - surfaces cross-domain links you weren't looking for |
| Write an article | /create-article <topic> - synthesizes your permanent notes into a draft |
| Get advice grounded in your thinking | /advise - any problem statement pulls relevant KB insights first |
| Extract insights from a document | /extract-document-insights <file> - processes papers, books, reports |
| Extract insights from your own content | /extract-insights <file> - processes your conversations, transcripts, notes |
| Deep research on a topic | /deep-research <topic> - discovers, extracts, and integrates new insights |
| Run a dialectic | /dialectic <question> - two committed positions stress-test your idea |
Abilities Marketplace
Cornelius supports installable plugins from the Ability AI marketplace - modular capability packs you can add without touching core agent config.
Available Plugins
| Plugin | What It Adds | Skills You Get |
| trinity-onboard | Deploy any agent to Trinity - sovereign, self-hosted AI infrastructure with cron scheduling, multi-agent orchestration, approval gates, and audit trails. Same agent runs locally (interactive dev) and remotely (scheduled/always-on), synced via Git. | /request-trinity-access, /trinity-onboard, /credential-sync, /trinity-sync, /trinity-remote, /trinity-schedules |
| playbook-builder | Create structured, transactional playbooks with state management - autonomous, gated, or manual execution modes | /create-playbook, /save-playbook, /adjust-playbook |
| dev-methodology | Documentation-driven development: 5-phase cycle (context → develop → test → docs → PR), 14 skills, 3 sub-agents | /dev-methodology:init, /read-docs, /implement #N, /validate-pr N |
| utilities | Ops workflows for SSH/Docker services: incident investigation, safe deploys, log management, conversation saving | /investigate-incident, /safe-deploy, /docker-ops, /save-conversation |
Trinity: Deploying Your Agent
Install first:
claude plugin add abilityai/abilities && claude plugin install trinity-onboard@abilityaiOnce installed, the workflow is: ACCESS → ADOPT → SYNC → SCHEDULE
/request-trinity-access- get credentials (self-host or managed by Ability AI)/trinity-onboard- create required files for Trinity compatibility/trinity-sync push- deploy agent state to remote via Git/trinity-schedules schedule my-skill "0 9 * * *"- schedule recurring autonomous runs
Installing a Plugin
From your terminal (recommended):
# First time - add the Ability AI marketplace
claude plugin add abilityai/abilities
# Then install a plugin
claude plugin install trinity-onboard@abilityai
claude plugin install playbook-builder@abilityai
claude plugin install dev-methodology@abilityai
claude plugin install utilities@abilityaiOr from inside a Claude Code session:
/plugin marketplace add abilityai/abilities
/plugin install trinity-onboard@abilityaiUpdating Plugins
# Update a specific plugin
claude plugin update trinity-onboard@abilityai
# Update all installed plugins
claude plugin update --allCheck What's Installed
ls ~/.claude/plugins/cache/abilityai/After Installing: First Command to Run
| Plugin | Run This First | What It Does |
| trinity-onboard | /request-trinity-access | Get credentials for a Trinity instance (self-host or managed) |
| playbook-builder | /create-playbook | Build your first structured, reusable workflow |
| dev-methodology | /dev-methodology:init | Initialize docs-driven dev cycle in your project |
| utilities | /save-conversation or /investigate-incident | Ops tools - use as needed |
Where to Go Next
- New to the vault? Start with
/search-vault <something you think about>to see what's already there - Want to capture something? Just describe your idea - Cornelius will detect if it's worth saving
- Want to write something? Try
/recall <topic>first to see what raw material exists, then/create-article - Want to explore? Run
/auto-discoveryfor a serendipitous tour of connections in your knowledge graph
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

