Vault Category Guide

This is the explanatory index of the second brain. Unlike a link-only graph, it describes what every category is for and what each note contributes. Use Vault Knowledge Graph when you want relationships; use this guide when you want to understand what to open.

flowchart TD I[Capture in Inbox] --> J[Reflect in Journal] J --> K[Distill into Knowledge] S[Study Sources] --> K K --> M[Organize with Maps] M --> P[Apply through Playbooks] P --> T[Start from Templates] T --> R[Deliver Projects] A[AI Assistant] --> P A --> R

Vault root — System entry points

The main product, design, and build specifications that orient both the owner and coding assistants.

Coding Second Brain

!tip Start here Capture quickly in 00 Inbox/Inbox, choose a workflow from 02 AI Assistant/AI Coding Assistant, and turn durable lessons into notes linked from 01 Maps/Coding Knowledge Map. - Inbox 00 Inbox/Inbo

Design System and Visual Direction

This is the durable visual specification for humans and AI coding assistants. It synthesizes the two supplied Figma files while preserving their two distinct design modes: 1. Austonian product UI — compact, hig

SecBrain — Full Product and Coding Prompt

!important This is the master implementation brief for a coding agent. Replace every DECISION REQUIRED value from the owner's answers before building or deploying. Build SecBrain , a mobile-first personal secon

00 Inbox — Capture

Temporary ideas, questions, errors, and links waiting to be processed into durable knowledge or project work.

Inbox

Capture first; organize later. Each item should eventually become a task, project note, durable concept, reference, or deletion. - 1. Is there an action? Move it to a project and write the next physical step. 2

01 Maps — Knowledge maps

High-level navigation and mental models that show how concepts, learning paths, and workflows relate. Start here when exploring a domain.

AI Engineering Map

Tokens and context windows - Embeddings and semantic search - Sampling, temperature, and structured output - Tool calling and agent loops - Retrieval-augmented generation - Evaluation datasets and graders - L

Career and Learning Map

Pick a real problem just beyond current ability. 2. Predict before searching. 3. Build the smallest working version. 4. Explain the mechanism in your own words. 5. Record mistakes in 04 Journal/Learning Log.

Coding Knowledge Map

Use this as a curriculum and navigation page. Create a focused note when a subject becomes relevant; do not study everything in order. - Programming models: imperative, functional, object-oriented, declarative

Full-Stack Developer Learning Path

This path combines the Burmese-language libraries of Ei Maung (https://eimaung.com) and Saturngod (https://www.saturngod.net) with the evidence-driven agent workflow in 02 AI Assistant/Spec Map Build Prove Lear

Software Delivery Map

Idea → requirements → design → implementation → review → test → release → observe → learn. - Problem statement and non-goals - Acceptance criteria and 02 AI Assistant/Definition of Done - Architecture decision

Vault Knowledge Graph

This map keeps every durable note reachable and shows how the vault moves from sources to concepts, practice, projects, and reflection. Individual notes also contain contextual links back into this graph. - Cod

02 AI Assistant — AI collaboration

Prompts, rules, context, and delivery methods that make an AI coding assistant reliable, scoped, and evidence-driven.

Agent Rules

Reusable instructions for an AI coding assistant: Add project commands, architectural boundaries, generated-file rules, and forbidden operations to the repository's own agent instruction file. Keep global prefe

AI Coding Assistant

The main operating system for this vault is SPEC → MAP → BUILD → PROVE → LEARN 02 AI Assistant/Spec Map Build Prove Learn. The developer's role is not merely code writer; it is system designer, direction giver,

Context Pack

Maintain this page as the concise profile you can paste into a new AI coding session. - Preferred name: - Current skill level: - Languages and frameworks: - Operating system and tools: - Current learning goals:

Definition of Done

Acceptance criteria are satisfied and non-goals remain untouched - Behavior is verified with relevant automated or manual checks - Error, empty, loading, boundary, and permission states were considered - Secu

Prompt Library

Replace bracketed fields. Good prompts describe outcomes and evidence; they do not micromanage every keystroke. For multi-step feature work, use the staged prompts in 02 AI Assistant/Spec Map Build Prove Learn

Saturngod-Inspired Repository Guide

A reusable pattern for a project-level AGENTS.md . Adapt it to facts in the repository and delete irrelevant sections. Precision matters more than length. 1. Declare the source of truth for product scope and ac

Spec Map Build Prove Learn

This is the default workflow for substantial AI-assisted coding. It replaces a fragile prompt → code → bug → more prompts spiral with bounded work and evidence. A usable specification answers: - WHO has the pro

03 Projects — Projects

Outcome-oriented workspaces connecting goals, next actions, decisions, implementation evidence, and lessons learned.

Example Project

What useful change exists when this is finished? Who has the problem, and why solve it now? - - - - - 05 Templates/Architecture Decision Record - 2026-08-12 — Project created. - Test plan: - Rollback plan: - Su

Projects Dashboard

Project Outcome Next action Status --- --- --- --- 03 Projects/Example Project Replace with a real outcome Write the problem statement status/seed - - - Every active project needs one visible next action. - Tra

04 Journal — Learning and reflection

Chronological learning records that capture what changed, why it matters, and which durable note or project should receive the insight.

Learning Log

Built the foundation of this coding second brain. - Next: personalize 02 AI Assistant/Context Pack and replace 03 Projects/Example Project with a real project. - 01 Maps/Vault Knowledge Graph - 04 Journal/Lea

05 Templates — Reusable templates

Structured starting points for decisions, investigations, workflows, daily notes, and product discovery. Copy these instead of starting from a blank page.

Most important outcome: - First concrete action: - - - Prediction: - Result: - Lesson: - Finished: - Blocked: - Next action for tomorrow: - 01 Maps/Vault Knowledge Graph - 02 AI Assistant/Spec Map Build Prove

ADR — Decision title

Proposed What forces and constraints make a decision necessary? - 1. Option: - Benefits: - Costs/risks: We will... - Positive: - Negative: - Follow-up: What evidence would show this decision is wrong or should

Agentic Workflow Canvas — {{title}}

User: - Goal: - Deliverable: - Non-goals: - Success metrics: - Unacceptable failures: - Simplest direct approach: - Current quality / latency / cost: - Why a workflow may help: Step Input Output contract Tool

Bug — Short symptom

Expected: Actual: - Exact error: - Environment/version: - First known occurrence: Hypothesis Evidence for Evidence against Next test --- --- --- --- - 01 Maps/Vault Knowledge Graph - 02 AI Assistant/Spec Map

Concept

Broader: - Narrower: - Similar/contrasting: - - 01 Maps/Vault Knowledge Graph - 02 AI Assistant/Spec Map Build Prove Learn - 03 Projects/Example Project - 05 Templates/Daily Note

Web Product Discovery

Use this before building or redesigning any website or web application. The AI assistant must ask the owner these questions before deciding navigation, visual language, product identity, or deployment naming. -

06 Playbooks — Engineering playbooks

Repeatable procedures for debugging, design, reviews, security, testing, performance, and agentic delivery. Use during real work.

Agentic AI Development Loop

Use this playbook to turn a candidate task into a reliable workflow. - Name the user, goal, inputs, deliverable, and non-goals. - Define observable success and unacceptable failure. - Decide whether a direct pr

Code Review

Review behavior before style. 1. Intent: does the change match the stated problem and scope? 2. Correctness: invariants, boundaries, state transitions, error paths, concurrency. 3. Security/privacy: trust bound

Debugging

State the symptom precisely: expected, actual, environment, frequency, impact. 2. Reproduce with the smallest deterministic case. 3. Find the boundary where correct state becomes incorrect. 4. Form ranked, f

Performance Investigation

Define the user-visible metric and target: latency percentile, throughput, memory, cost, or startup time. 2. Measure a representative baseline. 3. Profile to locate the dominant bottleneck. 4. Form a hypothe

Security Review

Inputs: validate type, size, format, origin, and authorization. - Identity: authenticate reliably; authorize every protected action and object. - Data: minimize collection, encrypt appropriately, define reten

System Design

Define users, core journeys, scale assumptions, and non-goals. 2. Rank quality attributes: availability, consistency, latency, durability, privacy, cost. 3. Model data and invariants. 4. Draw boundaries, API

Testing Strategy

Tests provide confidence about behavior at an acceptable maintenance cost. - Unit: pure logic, boundaries, transformations - Integration: databases, queues, files, APIs, framework wiring - Contract: compatibili

07 Knowledge — Evergreen knowledge

Technology-independent concepts and engineering principles distilled from sources and experience.

Agentic AI Real-World Applications

These projects apply the course patterns in increasing order of autonomy. Outcome: answer a technical question with claim-level citations and visible uncertainty. - Steps: clarify → plan queries → retrieve → ex

API Design

An API is a contract between independently changing systems. - Resources/actions and stable domain language - Explicit schemas and validation - Authentication versus per-resource authorization - Idempotency for

Architecture Principles

Make boundaries follow business capabilities and rates of change. - Keep dependencies pointing toward stable policy, away from volatile details. - Hide decisions likely to change behind clear interfaces. - Pr

Databases

Choose a data model from access patterns, invariants, consistency needs, and operational constraints—not fashion. - Schema and constraints make invalid states harder to represent. - Indexes exchange write/stora

Design Patterns Selection Guide

Design patterns are named arrangements of responsibilities and communication. They are useful when they directly address a recurring change or coordination problem. They are harmful when added merely to make co

Git and Version Control

Git stores a graph of snapshots. Branches are movable names pointing to commits; merges and rebases create different histories from the same changes. - Commit one coherent reason for change. - Inspect status an

Web Fundamentals

DNS resolves a host; a transport connection is established; TLS authenticates and encrypts; HTTP carries requests and responses; browsers parse resources, build page structures, lay out, paint, and run scripts.

08 Sources — Source library

Evidence, course syntheses, books, PDFs, repositories, and reference studies from which durable notes are derived.

Agentic Fullstack Development Webinar

Local PDF: /Users/kaungkhantko/Downloads/agentic fullstack development webinar.pptx (1).pdf - Author/presenter: Kyaw Ko Ko Tun, Let's Tech Club - Format reviewed: 29-slide presentation - Main derivative workf

How categories work together

Capture uncertain material in Inbox, then process it into a project, journal entry, or evergreen note.
Use sources as evidence, not as the final organization system. Distill reusable ideas through Coding Knowledge Map and organize navigation with Vault Knowledge Graph.
Apply knowledge through engineering playbooks and begin recurring work with reusable templates.
Keep outcomes and proof in Projects Dashboard.
Teach stable working preferences to the assistant through AI Coding Assistant.