OpenParable
Skip to content
← Back to skills

claude-mega-brain

Research

OKF knowledge navigation. ALWAYS use this skill the moment a `<mega-brain>` block appears in session context — it means the project has documented concepts and you must consult them before answering questions about data, schemas, metrics, APIs, or systems. Also trigger when the user asks about a table, column, metric definition, service, or API endpoint and there's any chance it's documented in the project. When in doubt, read the file directly.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/guhcostan/claude-mega-brain/blob/HEAD/skills/claude-mega-brain/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/claude-mega-brain, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

claude-mega-brain — OKF Knowledge Navigator

When <mega-brain> context is present, the project has OKF-documented concepts — any Markdown file with type: in its YAML frontmatter. These are the authoritative source for domain knowledge: exact schemas, metric formulas, API contracts, runbooks, business definitions. Answers from the knowledge base beat answers from training data.

What OKF files look like

Any .md file anywhere in the project with YAML frontmatter containing type::

---
type: BigQuery Table
title: Orders
description: One row per completed customer order.
resource: https://...
tags: [sales, revenue]
timestamp: 2026-06-29T00:00:00Z
---

# Schema
| Column | Type | Description |
...

No dedicated folder required. The file can live in docs/, wiki/, project root, anywhere.

Navigation

  1. Read the file — paths in <mega-brain> are relative to the project root; use the Read tool directly
  2. Follow links — [text](rel-path.md) links in the body point to related concepts; read them
  3. log.md — chronological changelog; last 3 entries are injected automatically
  4. Linked concepts auto-surface — after you read an OKF file, linked concept summaries appear via PostToolUse hook

When to proactively read files

  • User asks about a data source, table, column, or metric — check <mega-brain> index first
  • Before writing SQL, queries, or code that touches project data
  • When a business term is ambiguous — OKF has the canonical definition
  • When the user says "check the knowledge base", "what do we know about X", or "olha no okf"

If <mega-brain> context is missing

No OKF concepts found in the project yet. Suggest:

  • /mega-brain:init — creates index.md and log.md to get started
  • /mega-brain:migrate — scans existing docs and generates OKF files automatically