Skip to content
This repository was archived by the owner on Jul 13, 2026. It is now read-only.

DeepShell Version 1.3.0!

Choose a tag to compare

View all tags
@ashes00 ashes00 released this 25 Aug 23:34
· 18 commits to main since this release
1.3.0
b35b168

🎯 Overview

🚀 We're excited to announce DeepShell v1.3.0, a major feature release that significantly expands DeepShell's capabilities with OpenRouter.ai integration, secure configuration backup/migration, and numerous enhancements that make DeepShell the most comprehensive LLM command-line interface available.

✨ New Features

🌐 OpenRouter.ai Integration

Complete third LLM service with full feature parity

  • 200+ Model Access: Connect to OpenAI, Anthropic, Meta, Google, Mistral, and many more providers through a single interface
  • Advanced Model Browser:
    • Paginated model selection (15 models per page)
    • Smart sorting: Free models listed first, then alphabetical
    • Navigate with n (next), p (previous), q (quit)
  • Multi-Key Management: Store multiple OpenRouter API keys with custom nicknames
  • Full Feature Support: All DeepShell features work seamlessly with OpenRouter
  • Free Model Support: Easy access to free models like openai/gpt-4o-mini:free

🔐 Configuration Backup & Migration

Enterprise-grade configuration management

  • Encrypted Export (-b filename.config):
    • Password-protected with confirmation
    • Secure binary format (completely unreadable as text)
    • Saves to Downloads folder automatically
    • Includes ALL settings, API keys, and configurations
  • Secure Import (-c filename.config):
    • Password verification
    • Confirmation prompt before overwriting
    • Future-proof with version metadata
  • Perfect for:
    • Backing up your complete DeepShell setup
    • Migrating between development machines
    • Sharing team configurations securely
    • Disaster recovery

🔑 Enhanced API Key Management

Unified and powerful key management across all services

  • Service-Agnostic Commands:
    • -set-key: Now manages both Gemini AND OpenRouter keys
    • -show-key: Shows active key for current LLM service
    • -a / --active-config: Quick summary of active LLM, model, and API key
  • Consistent UX: OpenRouter key management now matches Gemini's interface exactly
  • Smart Validation: Both services support nickname-based multi-key workflows

🛠️ Improvements & Fixes

User Experience Enhancements

  • Alphabetized Help Menu: All command-line options now sorted alphabetically for easier reference
  • Updated Interactive Logo: Now properly shows "Multi-LLM Support (Ollama, Gemini, and OpenRouter)"
  • Improved Error Handling: Better error messages and validation throughout
  • Enhanced Model Selection: Improved pagination and user-friendly navigation

Technical Improvements

  • Auto-Setup Bypass: Export/import flags now correctly bypass automatic setup
  • Future-Proof Design: Export format includes version metadata for cross-version compatibility
  • Binary Security: Export files are true binary format, preventing accidental text viewing
  • Memory Management: Enhanced memory handling and cleanup throughout

Bug Fixes

  • Fixed import flag being overridden by default configuration setup
  • Resolved model selection issues for OpenRouter
  • Corrected API key management inconsistencies
  • Fixed various edge cases in configuration handling

📊 What's Included

Complete Configuration Export

When you export your configuration, everything is included:

  • All LLM service configurations (Ollama, Gemini, OpenRouter)
  • All API keys with their nicknames
  • Model selections for each service
  • Interactive settings (history limit, streaming, animation)
  • Markdown rendering preferences
  • Server addresses and site attribution

OpenRouter Model Categories

Access to major model families including:

  • OpenAI: GPT-4, GPT-3.5 (including free variants)
  • Anthropic: Claude 3.5 Sonnet, Claude 3 Haiku
  • Meta: Llama 3.1, Llama 3.2 (including free versions)
  • Google: Gemma 2, PaLM models
  • Mistral: Mixtral, Mistral 7B
  • And 190+ more models from various providers

🚀 Getting Started with New Features

Try OpenRouter

# Setup OpenRouter
./deepshell -s
# Select option 1 (Manage LLM Services)
# Choose 3 (OpenRouter)

# Quick model change with new pagination
./deepshell -m
# Browse models with n/p navigation, free models listed first

# Manage multiple OpenRouter keys
./deepshell -set-key
# Choose 2 (OpenRouter)

Backup Your Configuration

# Export your complete setup
./deepshell -b my-deepshell-backup.config
# Enter password twice for protection

# Import on another machine
./deepshell -c my-deepshell-backup.config
# Enter password and confirm overwrite

Quick Status Check

# See your current setup at a glance
./deepshell -a
# Shows: LLM Service, Model, API Key (with nickname)

🔧 Migration Guide

From v1.2.x

No breaking changes! Your existing configuration will work seamlessly. New features are additive.

Recommended Actions

  1. Backup First: ./deepshell -b v1-2-backup.config
  2. Try OpenRouter: Add it as a third LLM option
  3. Test New Commands: Explore -a for quick status checks

🎯 What's Next

DeepShell v1.3.0 establishes a solid foundation for multi-LLM management with enterprise-grade backup capabilities. Future releases will focus on:

  • Additional LLM service integrations
  • Advanced conversation management features
  • Enhanced interactive mode capabilities
  • Performance optimizations

📚 Documentation

Full documentation with examples and tutorials is available in our [README.md](README.md).

🙏 Acknowledgments

Special thanks to the community for feature requests, bug reports, and testing that made this release possible.

Download DeepShell v1.3.0 from the [releases page](https://github.com/ashes00/deepshell/releases) or build from source.

Need Help? Check our [README.md](README.md) or open an issue on GitHub.

Happy Querying! 🤖✨

Assets 6
Loading