cryptex-ai - v0.3.3: True Temporal Isolation Implementation

🔥 Major Fix - True Temporal Isolation

This release implements genuine temporal isolation where AI services receive placeholders while functions get real secrets through intelligent monkey-patching.

🚀 What's New

Core Fix: - ✅ Fixed @protect_secrets decorator to intercept actual AI library calls - ✅ AI services now receive {{PLACEHOLDERS}} while functions get real values - ✅ Context manager for temporary monkey-patching during function execution - ✅ Automatic cleanup and restoration of original methods

Test Infrastructure: - 🧪 Reorganized test suite into proper unit/integration/security/performance structure
- 🧪 135 comprehensive tests proving temporal isolation works - 🧪 Live demonstration scripts showing real vs placeholder data - 🧪 Comprehensive AI call interception validation

Documentation: - 📚 Updated README.md with accurate implementation examples - 📚 Fixed llms.txt to reflect monkey-patching architecture - 📚 All docstrings updated to show working examples - 📚 Added live test documentation

🎯 Core Value Delivered

One decorator line = complete temporal isolation

python @protect_secrets(["openai_key"]) async def ai_tool(prompt: str, api_key: str) -> str: # Function receives: real API key for processing # AI service receives: {{OPENAI_API_KEY}} (intercepted) import openai return await openai.chat.completions.create( model="gpt-4", messages=[{"role": "user", "content": prompt}], api_key=api_key # Real key for function, placeholder to AI )

🛡️ Security Model

Three-phase temporal isolation: 1. Sanitization: Convert secrets to placeholders for AI processing 2. AI Processing: AI receives safe placeholders via monkey-patching
3. Resolution: Functions get real values for execution

📊 Validation

✅ 135 tests passing including AI interception validation
✅ Live demonstration scripts prove concept works
✅ Zero dependencies, zero configuration required
✅ <5ms sanitization, <10ms resolution performance

This release finally delivers on Cryptex's core promise: bulletproof secrets isolation with zero cognitive overhead.

- Python
Published by AnthemFlynn 7 months ago

cryptex-ai - v0.3.2

Full Changelog: https://github.com/AnthemFlynn/cryptex-ai/compare/v0.3.1...v0.3.2

- Python
Published by github-actions[bot] 8 months ago

cryptex-ai - v0.3.1

Full Changelog: https://github.com/AnthemFlynn/cryptex-ai/compare/v0.3.0...v0.3.1

- Python
Published by github-actions[bot] 8 months ago

cryptex-ai - v0.3.0: Repository Migration & Infrastructure

Major Changes

Added

GitHub Issue Templates: Comprehensive bug report, feature request, and security issue templates
Release Automation: Pre-release and release workflows with automated PyPI publishing
Documentation Site: Full MkDocs-based documentation deployed to GitHub Pages
CI/CD Improvements: Consolidated main workflow with parallel job execution
Codecov Integration: Code coverage reporting with proper configuration

Changed

Repository Rename: Complete migration from cryptex to cryptex-ai across all files
Documentation Updates: Consistent naming and improved installation guides
Development Workflow: Clarified uv requirement in README development setup
CI Environment: Added CRYPTEX_SKIP_PERF_CHECKS for stable CI test runs

Fixed

CI/CD Pipeline: Fixed all workflow issues including dependency installation and test execution
Documentation Naming: Corrected all references to use cryptex-ai consistently
Build Configuration: Updated Makefile project name from codename to cryptex_ai

Documentation

Installation Guides: Changed "pip (recommended)" to "pip (default)" to reflect standard practice
Complete Rebranding: All documentation now uses consistent cryptex-ai naming
Enhanced Examples: Updated all example code to use correct import statements

Installation

Note: PyPI release pending trusted publishing configuration.

```bash

Will be available after PyPI configuration:

pip install cryptex-ai==0.3.0 ```

What's Next

The v0.3.0 release establishes the complete infrastructure for the cryptex-ai project: - Fully automated CI/CD pipeline - Comprehensive documentation site - Professional project management tools - Consistent branding and naming

Full Changelog

See the CHANGELOG.md for complete details.

- Python
Published by AnthemFlynn 8 months ago

cryptex-ai - v0.2.0: Zero-Config Architecture Revolution

🚀 Zero-Config Architecture Revolution

The Big Change

Cryptex v0.2.0 eliminates configuration files entirely! No more cryptex.toml, no more setup, no more friction. Just pure, bulletproof temporal isolation with zero cognitive overhead.

✨ What's New

🔧 Zero Configuration Required

```python

Works immediately - no setup needed!

@protecttool(secrets=["openaikey"]) async def aitool(prompt: str, apikey: str) -> str: # AI sees: "{{OPENAIAPIKEY}}" # Tool gets: real API key return await openaicall(prompt, apikey) ```

📦 Built-in Patterns (95% of users)

OpenAI Keys: sk-... → {{OPENAI_API_KEY}}
Anthropic Keys: sk-ant-... → {{ANTHROPIC_API_KEY}}
GitHub Tokens: ghp_... → {{GITHUB_TOKEN}}
File Paths: /Users/... → /{USER_HOME}/...
Database URLs: postgres://... → {{DATABASE_URL}}

🛠️ Custom Patterns (5% of users)

```python from cryptex.patterns import register_pattern

registerpattern("slacktoken", r"xoxb-[0-9-a-zA-Z]{51}", "{{SLACK_TOKEN}}") ```

📖 Professional Documentation

318-line comprehensive README following Python library best practices
Professional badge section, performance metrics, architecture diagrams
Framework integration examples for FastMCP and FastAPI
Contributing guidelines and development setup

🏗️ Technical Improvements

Eliminated Dependencies

Removed: tomli>=2.0.1 (no TOML parsing needed)
Result: Lighter, faster, more secure

Architectural Benefits

🚫 No Attack Surface: Zero config files to inject or exploit
⚡ Lightning Fast: No file I/O, zero parsing overhead
🎯 Middleware Focused: Lightweight, predictable, no external deps
🔒 Security First: All configuration in version-controlled code

Performance Metrics

| Metric | Performance | Context | |--------|-------------|---------| | Sanitization | <5ms | 1KB payloads | | Resolution | <10ms | 10 placeholders | | Memory Overhead | <5% | vs unprotected apps | | Startup Time | 0ms | No config loading |

🔄 Migration Guide

✅ No Migration Needed (95% of users)

Your existing decorators work unchanged: python @protect_tool(secrets=["openai_key"]) # Still works perfectly\!

🗑️ Remove Config Files

Delete cryptex.toml files (no longer needed)
Remove environment variables (optional)

🛠️ Custom Patterns (if you had any)

Replace TOML config with registration API: ```python

Before: cryptex.toml

[secrets]

patterns = [{name = "my_key", pattern = "...", template = "..."}]

After: Python code

from cryptex.patterns import registerpattern registerpattern("my_key", r"pattern", "{{TEMPLATE}}") ```

🔮 What This Means

This isn't just an update - it's a paradigm shift. Cryptex now delivers on its core promise:

"One decorator line = complete temporal isolation"

No setup, no config, no friction. Just bulletproof secrets isolation that works immediately.

📊 Release Impact

| Metric | v0.1.0 | v0.2.0 | Change | |--------|--------|--------|---------| | Config Required | Yes | No | ✅ Zero config | | Dependencies | 2 | 1 | ✅ 50% reduction | | Setup Time | ~10ms | 0ms | ✅ Instant | | Attack Surface | Config files | None | ✅ Eliminated | | Lines of Code | 1,400+ | 1,128 | ✅ Simpler |

🙏 Thank You

This release represents months of architectural thinking and represents the future of middleware libraries - zero configuration, maximum security, perfect developer experience.

Made with ❤️ for the AI/LLM community

Full Changelog: https://github.com/AnthemFlynn/cryptex/compare/v0.1.0...v0.2.0