# Memory-Assisted Shaping

Name: Memory-Assisted Shaping
Type: AI continuity protocol
URL: https://xxyoudeadpunkxx.github.io/Memory-Assisted-Shaping/
Repository: https://github.com/XxYouDeaDPunKxX/Memory-Assisted-Shaping
Runtime: ChatGPT Projects with append-only Python note persistence
License: CC BY-SA 4.0
License URL: https://creativecommons.org/licenses/by-sa/4.0/
Canonical files: protocol.md, note-io.md, notes.py

## What this is

Memory-Assisted Shaping is a ChatGPT Projects protocol for shaping one idea across a long conversation without losing decisions, gates, discarded paths, or source boundaries.

It is not a full note-taking system. It is a continuity protocol that keeps shaping state separate from the final artifact.

## Operating line

GPT shapes. Python persists. Memory stays append-only. The final artifact is produced only after explicit approval.

## Use case

Use Memory-Assisted Shaping when one idea needs a long conversation and the session must preserve decisions, open gates, discarded paths, source boundaries, and artifact readiness without turning notes into the main work surface.

## Canonical source

The canonical project files are:

- protocol.md
- note-io.md
- notes.py

Treat the repository as the live public source. Treat this file as the LLM-readable project index, not as the full protocol specification.

## Operating model

GPT shapes the idea before producing an artifact.

When continuity would degrade without retention, GPT emits a minimal memory signal.

Python timestamps and appends the signal.

GPT does not own, normalize, or rewrite memory files.

The final artifact is produced only after explicit approval.

## Tracked state

- decisions
- open gates
- discarded paths
- source boundaries
- artifact readiness

## Live signal types

- mode
- shape
- gate
- decision
- discard
- tension
- source
- artifact

## Repository files

- protocol.md: behavioral protocol GPT should follow while shaping one active idea.
- note-io.md: boundary between GPT signals and Python persistence.
- notes.py: Python helper that appends internal notes.

## Generated session files

- session_meta.json
- session_notes.jsonl
- session_export.json

## Usage summary

Add the project files to ChatGPT Project Source.

Tell GPT: Read protocol.md and use Memory-Assisted Shaping for this session.

Ask for a recap when state matters.

Ask for a final artifact only when the shape is ready.

## Discovery Set

Machine-readable project files:

- llms.txt: LLM-readable project index.
- raw-manifest.json: structured machine-readable project manifest.
- robots.txt: crawler access declaration.
- sitemap.xml: sitemap for the GitHub Pages surface.

HTML discovery signals:

- canonical URL
- rel=alternate for llms.txt
- rel=alternate for raw-manifest.json
- JSON-LD SoftwareSourceCode metadata
- Open Graph metadata
- footer machine-readable links

## Keywords

AI protocol, ChatGPT Projects, long session continuity, idea shaping, append-only notes, source boundaries, Memory-Assisted Shaping, AI continuity protocol, artifact readiness, session drift control