AiContext

A personal toolkit for working smarter with AI — structured context, dense stamps, and portable session state..

currently just a preliminary assessment that is ideally:

based on a wiki Foundation, AiH augmented, exploiting unicodeSymbols and nonWord stringPatterns. triggered via TwH( GitCalTwe ).

.

refined '0 QuickLinks`

---

What Is This Repo?

AI conversations are stateless. Every new session starts blank.
AiContext is a collection of tools and documents that solve that problem — capturing, compressing, and injecting context so AI picks up exactly where you left off.

The philosophy: "denser is better". A well-structured JSON stamp beats a wall of prose every time.

---

Projects

ContextLatticeStamp/

The core system. Build versioned JSON stamps of your working context, inject them into any AI chat via hotkey, and maintain continuity across sessions.

What	Where
Docs & guides	ContextLatticeStamp/Docs/
Scripts (Python + AHK)	ContextLatticeStamp/Scripts/
Example stamps	ContextLatticeStamp/Examples/
Design essay	ContextLatticeStamp/Essay/

---

Repo Conventions

Naming: PascalCase for all directories and documents.
Private data: Files prefixed RnHw_ are personal/local only — never committed. Excluded via .gitignore.
Versioning: Long-form drafts use _vX.Y suffix (e.g. DenserIsBetter_v0.1.md).

See ContextLatticeStamp/Docs/FolderStructureGuide.md for the full layout and decision rules.

• also:
  • 2026-05-26: considering EndianNess
  • 2026-05-11: updates & 2026-05-26 ( offline )

---

What's Not Here

Excluded	Reason
RnHw_*.json	Personal vocabulary & private data dictionaries — local only
Session logs	Transient; not versioned

• 2026-06-13: pulseTweaks
• 2026-04-25: Jdate macro V0.0.1

---

Status

Active personal project. Structure is stable; tooling and documentation are expanding.

• confirm Sync(s) are active and effective
• Minimizing edits on this page reduces disruptions - leading to a "QualityCascade" (but AiContext fixed)

Plans:

... merge ...

---

TwH (markdown list):

• 🧩 Context Space
  • Current 🧩
    • TbH
      • Workflow Dispatcher
        • Menus
        • Launch Points
        • Context Switches
        • Resume Points
    • Target 🧩
      • Active Work ReView
• Notion interfaces/integrations
• boxChars FAILed

---

Background:

About Your AiContext Repo 🙂 Here's what I can see from the webpage, Hans:

Tech Stack Overview Languages * Python — 77.4% (dominant) * HTML — 17.1% * AutoHotkey (AHK) — 5.5%

Key Files & Their Roles

File	Purpose
jump_server.py	A local "JumpH" server
reg_url.py	URL registration utility
rnhw_json_manager.py	JSON data management
sync_datadict_to_twh.py	Data dictionary sync
Register.ahk	AutoHotkey hotkey triggers

Infrastructure * GitHub Actions — there's a .github/workflows directory with a Python application workflow, so CI is wired up * JSON — RnHw_DataDict.json appears to be a core data structure (personal vocabulary/private data)

The Core Concept The system is built around ContextLatticeStamp — essentially versioned JSON "stamps" that capture AI session state, injected via hotkey (AHK) to restore context across stateless AI conversations.

The philosophy is "denser is better" — compact structured data over verbose prose.

---

By HwWobbe