Skip to main content

AI Memory & Context Archivist - A framework for managing AI context and project memory

Project description

Proust Framework

"In reality, every reader is, while reading, the reader of his own self." — Marcel Proust

An AI memory and context management framework for software development, inspired by Marcel Proust's revolutionary exploration of memory, time, and narrative coherence.

What is Proust?

Proust transforms AI-assisted development from scattered interactions into coherent narrative experiences. Like its namesake's In Search of Lost Time, the framework helps AI assistants maintain deep contextual memory, creating development experiences where every action builds on accumulated understanding.

The Trinity: Claude + Simone + Proust work together as an integrated development experience where Claude provides the AI intelligence, Simone structures the project management workflow, and Proust maintains the persistent memory and context that connects everything across time.

Personas

Proust — The Memory & Context Archivist

Keeps long-term knowledge, cross-project recall, and narrative coherence. Named after Marcel Proust, whose literary exploration of involuntary memory revolutionized our understanding of time and consciousness. Proust the AI persona maintains contextual awareness across sessions, ensuring every development decision fits into the larger story.

Installation

Install the Proust Framework using pip or uv:

# Using uv (recommended)
uv tool install proust-framework
proust install

# Using pip
pip install proust-framework
proust install

# Or run without installing
uv run --with proust-framework proust install

Quick Start

  1. Install Framework: Use uv tool install proust-framework
  2. Customize Templates: Follow the customization steps below
  3. Start Using: Initialize your first project

Customization

Step 1: Replace Template Placeholders

All templates use this placeholder format:

{{PLACEHOLDER_NAME}}

Examples:

  • {{PROJECT_NAME}} → Replace with your actual project name
  • {{YYYY-MM-DD}} → Replace with current date in ISO format
  • {{COMPANY_NAME}} → Replace with your organization name

Step 2: Key Files to Customize

Work through templates in this order:

1. Project Identity

  • .proust/ethos.md - Mission, vision, principles
  • .proust/manifesto/manifesto.yml - Technical architecture

2. Configuration Files

  • .proust/universal_claude.md - Claude Code settings
  • .proust/guardrails.yml - AI coding standards
  • .proust/brand.yml - Design system

3. Project Structure

  • .simone/00_PROJECT_MANIFEST.md - Project status
  • Templates in .simone/99_TEMPLATES/

Step 3: Use Examples for Reference

Check .examples/ directory for:

  • Complete framework configurations
  • All placeholders filled with example content
  • Production-ready memory patterns

Step 4: Remove Template Instructions

After customization:

  • Delete instruction blocks at top of files
  • Remove lines starting with "# HOW TO USE THIS TEMPLATE"
  • Clean up any remaining {{PLACEHOLDER}} references

Core Philosophy

Memory as Foundation

Context persists across sessions and projects, creating continuity where traditional development tools create fragmentation.

Narrative Coherence

Every change fits into the larger story, ensuring development decisions build on accumulated understanding rather than starting fresh each time.

Time and Reflection

Past decisions inform present actions through structured memory patterns, creating deeper, more thoughtful development.

Deep Context

Surface details reveal deeper patterns through persistent contextual awareness.

Framework Structure

.proust/ - AI Memory System

  • Memory Configuration - How AI assistants remember and recall context
  • Behavioral Patterns - Consistent AI behavior across sessions
  • Identity Preservation - Brand and voice consistency over time
  • Knowledge Architecture - Structured information organization

.simone/ - Project Management System

Built on the Simone project management framework by Helmi:

  • Milestones - Major project phases (M01, M02...)
  • Sprints - Focused work periods within milestones (S01, S02...)
  • Tasks - Atomic units of work (T01, T02...)
  • Documentation - Requirements, architecture, and decisions
  • Templates - Standardized formats for consistency

Directory Structure:

src/
├── .proust/                 # Memory Archive
│   ├── ethos.md            # Project identity
│   ├── universal_claude.md # AI protocols
│   ├── guardrails.yml      # Quality standards
│   ├── brand.yml           # Design system
│   └── commands/simone/    # Structured workflows
└── .simone/                # Project Structure
    ├── 00_PROJECT_MANIFEST.md
    ├── 01_PROJECT_DOCS/
    ├── 02_REQUIREMENTS/
    ├── 03_SPRINTS/
    ├── 04_GENERAL_TASKS/
    └── 99_TEMPLATES/

Key Features

  • 🧠 Persistent AI Memory - Context survives session boundaries
  • 📖 Narrative Development - Projects as evolving stories
  • 🔍 Memory Quality Assurance - Continuous narrative validation
  • 📚 Rich Context Patterns - Deep contextual understanding
  • Instant Memory Recall - AI assistants remember everything
  • 🎯 Coherent Decision Making - Every choice fits the larger narrative

Usage

Initialize New Project

  1. Install the framework using installation method above
  2. Customize core files (ethos.md, guardrails.yml, brand.yml)
  3. Create your first project manifest in .simone/00_PROJECT_MANIFEST.md
  4. Start using commands for structured workflows

Daily Development

  • Use commands from .proust/commands/simone/ for structured workflows
  • Maintain project context through the .simone/ structure
  • Let AI assistants access rich context for better development

Quality Assurance

  • Regular project reviews maintain narrative coherence
  • Built-in validation ensures framework integrity
  • Memory patterns persist across team members and sessions

Advanced Documentation

Use Cases

Individual Developers

  • Maintain project context across work sessions
  • Build consistent coding patterns and practices
  • Create coherent project narratives
  • Develop personal AI assistant memory

Development Teams

  • Share collective project memory
  • Maintain consistent team practices
  • Build institutional knowledge
  • Create team-specific AI behavioral patterns

AI Assistant Integration

  • Configure Claude Code with persistent memory
  • Establish consistent AI behavior patterns
  • Create project-specific AI capabilities
  • Build contextual AI workflows

Philosophy in Practice

"We don't receive wisdom; we must discover it for ourselves after a journey that no one can take for us or spare us." - Marcel Proust

The Proust Framework applies this wisdom to software development:

  • Discovery over Repetition - Each project builds understanding
  • Journey over Destination - Process creates lasting value
  • Memory over Information - Context becomes knowledge
  • Narrative over Tasks - Stories create meaning

Contributing

The Proust Framework evolves through community memory and narrative. Contributions should:

  1. Honor the philosophy - Maintain memory and narrative focus
  2. Build on existing patterns - Extend rather than replace
  3. Document thoroughly - Add to the collective memory
  4. Test comprehensively - Ensure narrative coherence

License

MIT License - Build your own memories and narratives.

Includes portions of the Simone project management framework by Helmi. See ACKNOWLEDGMENTS.md for full attribution.


Version: 0.3.2 Inspired by: Marcel Proust's In Search of Lost Time Built for: AI-assisted development with persistent memory Repository: https://github.com/renkasiyas/proust.git Package: https://pypi.org/project/proust-framework/

In memory of Marcel Proust (1871-1922), whose exploration of involuntary memory showed us that the past is never truly lost, only waiting to be rediscovered.

(Of course this was made with 🤖 Claude's help)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

proust_framework-0.3.2.tar.gz (54.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

proust_framework-0.3.2-py3-none-any.whl (10.7 kB view details)

Uploaded Python 3

File details

Details for the file proust_framework-0.3.2.tar.gz.

File metadata

  • Download URL: proust_framework-0.3.2.tar.gz
  • Upload date:
  • Size: 54.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.7.19

File hashes

Hashes for proust_framework-0.3.2.tar.gz
Algorithm Hash digest
SHA256 c1cb6b8748722caa5619dfabf012401bcd4a10cd5a9c2f5a89c3ccd6c9758103
MD5 bf29cebefcae4d50fb2aedbad9a242eb
BLAKE2b-256 aa046229bfa2b9cf1d552cc9a02f8a8ff5a392ef8137ae5b29d35985104f5924

See more details on using hashes here.

File details

Details for the file proust_framework-0.3.2-py3-none-any.whl.

File metadata

File hashes

Hashes for proust_framework-0.3.2-py3-none-any.whl
Algorithm Hash digest
SHA256 fe889ecbf770ba62966a9f9e22c28e1485503e986477798b34566a2bba8bdf4a
MD5 f0e5c5188f35c281c9a6247c1af5eebd
BLAKE2b-256 8c6d3c4cee36727d8a04c5d3f2b2a05b600643f6da5ac3dedf9ff93aa8889117

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page