Skip to main content

AI-assisted development context management tool for maintaining consistency and productivity

Project description

Quaestor

๐Ÿ›๏ธ Context management for AI-assisted development

PyPI Version Python 3.12+ License: MIT

Quaestor provides intelligent context management and quality enforcement for AI assistants, with flexible modes for personal and team projects.

Why Quaestor?

AI assistants like Claude are powerful but need context. Quaestor provides:

  • ๐Ÿง  Smart Context - Automatically adjusts rules based on project complexity
  • ๐ŸŽฏ Flexible Modes - Personal mode for solo work, team mode for collaboration
  • โš™๏ธ Command Customization - Override and configure commands per project
  • ๐Ÿ“Š Progress Tracking - Maintain project memory and milestones
  • โœ… Quality Enforcement - Ambient rules that work outside commands

Quick Start

# Personal mode (default) - Everything local to your project
uvx quaestor init

# Team mode - Shared commands, committed rules
uvx quaestor init --mode team

Personal Mode (Default)

Creates a self-contained setup in your project:

project/
โ”œโ”€โ”€ .claude/           # All AI files (gitignored)
โ”‚   โ”œโ”€โ”€ CLAUDE.md     # Context-aware rules
โ”‚   โ”œโ”€โ”€ commands/     # Local commands
โ”‚   โ””โ”€โ”€ settings.json # Hooks
โ””โ”€โ”€ .quaestor/        # Architecture & memory

Team Mode

For shared projects with consistent standards:

project/
โ”œโ”€โ”€ CLAUDE.md         # Team rules (committed)
โ”œโ”€โ”€ .quaestor/        # Shared documentation
โ””โ”€โ”€ ~/.claude/        # Global commands

Now Claude can use commands with project-specific behavior:

/task: implement user authentication
/status
/configure

Installation

# No install needed (recommended)
uvx quaestor init

# Or install globally
pip install quaestor

Commands

CLI Commands:

  • quaestor init - Initialize with smart defaults
    • --mode personal (default) - Local, self-contained setup
    • --mode team - Shared commands and rules
    • --contextual (default) - Analyze project complexity
  • quaestor configure - Customize command behavior
    • --init - Create command configuration
    • --command <name> --create-override - Override specific commands
  • quaestor update - Update while preserving your changes

AI Assistant Commands:

  • /task - Implement features with quality checks
  • /status - Show project progress
  • /help - List all commands
  • /milestone - Manage project phases
  • /check - Run quality validation
  • /auto-commit - Commit completed TODOs
  • /milestone-pr - Create PR for milestones

Key Features

๐Ÿง  Context-Aware Rules

Quaestor analyzes your project and applies appropriate rules:

  • Simple projects โ†’ Basic quality checks
  • Standard projects โ†’ Progressive workflow suggestions
  • Complex/Team projects โ†’ Strict enforcement

Rules work ambiently in CLAUDE.md, not just in commands!

โš™๏ธ Command Customization

Configure commands per project with .quaestor/command-config.yaml:

commands:
  task:
    enforcement: strict
    parameters:
      minimum_test_coverage: 90
      max_function_lines: 30
    custom_rules:
      - "All APIs must have OpenAPI specs"
      - "Database changes require migrations"

Or create full overrides in .quaestor/commands/task.md.

๐ŸŽฏ Flexible Modes

Personal Mode (Default):

  • Everything local in .claude/
  • Perfect for personal projects
  • Commands and context in one place
  • Fully gitignored

Team Mode:

  • Shared standards in .quaestor/
  • Global commands in ~/.claude/
  • Consistent across team
  • Version controlled rules

๐Ÿ“Š Smart Project Analysis

  • Auto-detects language (Python, Rust, JS/TS, Go, Java, etc.)
  • Identifies test frameworks and CI/CD
  • Recognizes team markers (CODEOWNERS, PR templates)
  • Calculates complexity score

๐Ÿ”„ Intelligent Workflow

Progressive enforcement based on context:

  • Simple tasks โ†’ Direct implementation
  • Complex tasks โ†’ Research โ†’ Plan โ†’ Implement
  • Multi-file changes โ†’ Automatic agent delegation
  • Continuous validation after every few edits

Project Structure

Personal Mode (Default)

your-project/
โ”œโ”€โ”€ .claude/                    # All AI files (gitignored)
โ”‚   โ”œโ”€โ”€ CLAUDE.md              # Context-aware rules
โ”‚   โ”œโ”€โ”€ commands/              # Local command copies
โ”‚   โ”‚   โ”œโ”€โ”€ task.md
โ”‚   โ”‚   โ”œโ”€โ”€ status.md
โ”‚   โ”‚   โ””โ”€โ”€ ...
โ”‚   โ””โ”€โ”€ settings.json          # Hooks configuration
โ”œโ”€โ”€ .quaestor/                 # Optional, for docs
โ”‚   โ”œโ”€โ”€ ARCHITECTURE.md        # Project structure
โ”‚   โ”œโ”€โ”€ MEMORY.md             # Progress tracking
โ”‚   โ”œโ”€โ”€ command-config.yaml   # Command customization
โ”‚   โ””โ”€โ”€ commands/             # Command overrides
โ”‚       โ””โ”€โ”€ task.md          # Custom task command
โ””โ”€โ”€ .gitignore                # Auto-updated

Team Mode

your-project/
โ”œโ”€โ”€ CLAUDE.md                  # Team rules (committed)
โ”œโ”€โ”€ .quaestor/                 # Shared documentation
โ”‚   โ”œโ”€โ”€ QUAESTOR_CLAUDE.md    # AI instructions
โ”‚   โ”œโ”€โ”€ CRITICAL_RULES.md     # Quality standards
โ”‚   โ”œโ”€โ”€ ARCHITECTURE.md       # Project structure
โ”‚   โ”œโ”€โ”€ MEMORY.md            # Progress tracking
โ”‚   โ”œโ”€โ”€ command-config.yaml  # Command config
โ”‚   โ””โ”€โ”€ hooks/               # Automation scripts
โ”œโ”€โ”€ ~/.claude/commands/       # Global commands
โ””โ”€โ”€ .claude/settings.json    # Local hooks only

How It Works

  1. Project Analysis - Scans for language, tests, complexity
  2. Context Generation - Creates appropriate CLAUDE.md rules
  3. Command Setup - Installs commands (local or global)
  4. Customization - Allows per-project overrides
  5. Smart Updates - Preserves your changes

Example Workflows

Simple Project (Personal Mode):

You: /task: add config parser

Claude thinks: "Simple project, basic rules"
- Writes clean code with tests
- Follows language idioms
- Updates documentation

Complex Project (Team Mode):

You: /task: refactor authentication system

Claude thinks: "Complex task, strict rules apply"
1. "I'll research the current auth implementation..."
2. "Here's my plan: [detailed plan]"
3. "I'll spawn agents for parallel work"
4. Validates every 3 edits
5. Updates milestone tracking

Command Customization Example

Create project-specific rules:

quaestor configure --init

Edit .quaestor/command-config.yaml:

commands:
  task:
    enforcement: strict
    custom_rules:
      - "All endpoints must have rate limiting"
      - "Use dependency injection pattern"

Now /task enforces your project standards!

Automated Hooks

Optional hooks enforce quality automatically:

  • Pre-edit - Ensure research before changes
  • Post-edit - Format code, update progress
  • Pre-commit - Run tests and quality checks
  • Milestone - Track progress, create PRs

Configure in .claude/settings.json (created during init).

Ambient Rule Enforcement

Unlike command-only systems, Quaestor's rules work everywhere:

<!-- In your CLAUDE.md -->
## ๐Ÿง  THINKING PATTERNS

Before EVERY response, I'll consider:
1. **Complexity Check**: 
   - Simple request? โ†’ Direct implementation
   - Multiple components? โ†’ "Let me research and plan this"
   
2. **Delegation Triggers**:
   if (files_to_modify > 3) {
     say("I'll spawn agents to handle this efficiently")
   }

Claude follows these patterns even outside /task commands!

Updating

# Check what would change
quaestor update --check

# Update with backup
quaestor update --backup

# Force update all files
quaestor update --force

Updates preserve your customizations in user-editable files.

Contributing

git clone https://github.com/jeanluciano/quaestor.git
cd quaestor
uv sync
uv run pytest

License

MIT


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

quaestor-0.4.0.tar.gz (112.5 kB view details)

Uploaded Source

Built Distribution

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

quaestor-0.4.0-py3-none-any.whl (126.1 kB view details)

Uploaded Python 3

File details

Details for the file quaestor-0.4.0.tar.gz.

File metadata

  • Download URL: quaestor-0.4.0.tar.gz
  • Upload date:
  • Size: 112.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.12.9

File hashes

Hashes for quaestor-0.4.0.tar.gz
Algorithm Hash digest
SHA256 44d2d2f9c106cc56b5885c53cd86c7e6f3c036340dcf0aec544e9dc6dadd84e0
MD5 f7b9da0e776d9981e27c6a636167d8f7
BLAKE2b-256 f0812a3d0a4e3b8533c4fadd74185498d6b602827cafa4894737b6253b7a0072

See more details on using hashes here.

Provenance

The following attestation bundles were made for quaestor-0.4.0.tar.gz:

Publisher: publish.yml on jeanluciano/quaestor

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file quaestor-0.4.0-py3-none-any.whl.

File metadata

  • Download URL: quaestor-0.4.0-py3-none-any.whl
  • Upload date:
  • Size: 126.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.12.9

File hashes

Hashes for quaestor-0.4.0-py3-none-any.whl
Algorithm Hash digest
SHA256 3cef68a1346d41904346c7696ca8dffb4ab35fdfc0e86df0b802ddf82cc3fc90
MD5 948d6bca072ca5b6ea1a47c6e555b230
BLAKE2b-256 9e4fe074c9068bec3394dd141724873749d7d7aaa3289740d06e09aefe149e02

See more details on using hashes here.

Provenance

The following attestation bundles were made for quaestor-0.4.0-py3-none-any.whl:

Publisher: publish.yml on jeanluciano/quaestor

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

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