Releases: Chat2AnyLLM/code-assistant-manager
Releases · Chat2AnyLLM/code-assistant-manager
1.3.0
What's Changed
- docs: Update CHANGELOG for v1.1.0 release by @zhujian0805 in #36
- feat: implement dynamic version detection from git tags by @zhujian0805 in #37
- feat: implement unified fetching framework and eliminate code duplication by @zhujian0805 in #38
- add skill support for Qwen code by @zhujian0805 in #39
- adding new features by @zhujian0805 in #42
Full Changelog: 1.1.0...1.3.0
v1.2.0
Release version 1.2.0 - **Fancy Name Generation**: Auto-generate creative prompt names like "Cosmic Coder" or "Quantum Assistant" when adding or importing prompts - **Enhanced Prompt Update Command**: New `cam prompt update` command with `--file` option to update prompt content, description, name, and default status - **Improved Prompt Status**: `cam prompt status` now shows file paths alongside app:level information for better visibility - **Optional Prompt Names**: `cam prompt add` and `cam prompt import` commands now support optional names with automatic fancy name generation - **Wildcard Configuration Support**: Added `*` wildcard support in `cam config show` for flexible pattern matching (e.g., `cam config show "claude.*.*.lastToolDuration"`) - **Prompt CLI Commands**: Updated command structure to be more intuitive: - `cam prompt add` - Add prompts (auto-generates names if not provided) - `cam prompt update` - Update existing prompts with content from files - `cam prompt import` - Import from live app files with fancy names - `cam prompt install` - Install configured prompts to app files - `cam prompt status` - Enhanced to show file paths
Release v1.1.0
What's New in v1.1.0
Added
- Static Model List Support: Add support for static model lists for endpoints, improving model discovery and configuration
- CLI Startup Optimization: Implement lazy loading for CLI startup to improve performance and reduce initialization time
- Multi-Source Repository Configuration: Add support for configuring repositories from multiple sources with parallel processing
- Multi-Model Selection: Add multi-model selection support for Goose/Codex/Droid/Continue agents
- Agent Metadata Pulling: Implement agent metadata pulling using awesome-claude-agents approach for better agent discovery
- Plugin Marketplace: Add support for thedotmack plugin marketplace with enhanced marketplace:plugin naming
- New Tool Support:
- Block Goose CLI tool support
- OpenCode MCP server and prompt support
- Enhanced engine type determination for various AI tools
- URL Display: Add URL display to 'extensions browse' command for better visibility
- Blackbox Integration: Add blackbox integration and documentation
- Tool Arguments: Pass tool arguments through CLI and show complete command with parameters
Changed
- CLI Refactoring:
- Rename 'extension' command to 'extensions' and update auto-complete
- Standardize CRUD patterns across CLI commands
- Fix MCP nesting issues
- Deprecate 'cam plugin browse' command in favor of enhanced 'list' command
- Engine Type Detection: Make engine type determination dynamic instead of hardcoded, with better OpenAI-compatible API support
- Model Configuration: Unify model listing logic into a single v1_models module
- Goose Menu: Combine goose menu into single unified list
Fixed
- Security Vulnerabilities: Implement security fixes for critical vulnerabilities
- Test Suite Issues: Resolve failing tests in CLI integration and plugin commands
- Plugin Conflicts: Implement plugin conflict resolution
- Configuration Issues: Multiple fixes for profile selection, MCP configurations, and config handling
- Installation Issues: Fix issue where cam update deletes npm packages
- Runtime Issues: Fix skill and agent repository loading with parallel processing
Security
- Critical Vulnerability Fixes: Addressed multiple security issues identified in security audit
v1.0.3
Merge pull request #10 from zhujian0805/main main