Releases: ThomasRohde/klondike-spec-cli
Releases · ThomasRohde/klondike-spec-cli
v0.2.11 - Feature Prompt Command
✨ Added
- \klondike feature prompt \ - Generate copilot-ready prompts for implementing features
- Includes feature description, acceptance criteria, and project context
- Pre-commit verification instructions (lint, format, test)
- Klondike workflow steps for feature completion
- --output \ option to write prompt to a file
- --interactive\ flag to launch GitHub Copilot CLI with the prompt
0.3.0: docs: improve copilot instructions based on session learnings
Fixed minor issues
v0.2.0
v0.2.0 - MCP Server Support"
v0.1.0 - Initial Release
Changelog
All notable changes to Klondike Spec CLI will be documented in this file.
The format is based on Keep a Changelog,
and this project adheres to Semantic Versioning.
0.1.0 - 2025-12-07
🎉 Initial Release
The CLI that built itself — This release represents the complete implementation of Klondike Spec CLI, developed across 4 AI coding sessions using the very methodology the tool implements.
✨ Added
Core Commands
klondike init [name]- Initialize .klondike directory with project artifactsklondike status- Display project status with progress bar, git info, and next prioritiesklondike validate- Check artifact integrity and consistencyklondike progress- Regenerate agent-progress.md from JSON data
Feature Management
klondike feature add- Add features with description, category, priority, and acceptance criteriaklondike feature list- List features with optional status filtering and JSON outputklondike feature show <id>- Display full feature details including verification evidenceklondike feature start <id>- Mark a feature as in-progressklondike feature verify <id>- Mark a feature as verified with evidenceklondike feature block <id>- Mark a feature as blocked with reasonklondike feature edit <id>- Edit feature notes, priority, and criteria (description immutable)
Session Management
klondike session start- Begin a coding session with focus trackingklondike session end- End session with summary, completed items, and handoff notes- Automatic git status integration shows uncommitted changes and recent commits
Reporting & Export
klondike report- Generate status reports in markdown, plain text, or JSONklondike export-features <file>- Export features to YAML or JSON with status filteringklondike import-features <file>- Import features from YAML or JSON with duplicate detection
Developer Experience
klondike completion <shell>- Generate shell completion scripts for bash, zsh, and PowerShell- Rich terminal output with colors, progress bars, and emoji icons
- JSON output mode for programmatic integration
Quality & Reliability
- Comprehensive input validation and sanitization
- Git integration for status tracking and commit logging
- Performance-optimized O(1) feature lookups via indexed data structures
- 98 tests with 74% code coverage
🔧 Technical Stack
- Python 3.10+ with full type annotations
- Pith (pypith) - Agent-native CLI framework with progressive discovery
- Rich - Beautiful terminal formatting
- PyYAML - Configuration and import/export support
- Hatchling - Modern Python build backend
📖 Documentation
- Comprehensive README with story-driven narrative
- Complete command reference with examples
- GitHub Actions CI/CD for testing and PyPI publishing
- MIT License
Development Journey
This project was developed in 4 AI coding sessions:
| Session | Focus | Features Added |
|---|---|---|
| 1 | Project foundation | Core models, init, status, feature CRUD |
| 2 | Session management | Session start/end, progress regeneration |
| 3 | Validation & reporting | Validate command, report generation |
| 4 | Polish & complete | Git integration, completion, import/export, performance |
Final stats:
- 30 features specified and verified
- 98 tests passing
- 74% code coverage
- 1600+ lines of Python
The Meta-Achievement
This CLI was built by an AI agent following the Klondike Spec methodology — tracking its own features, managing its own sessions, and providing verification evidence for each completed feature.
Built with 🤖 by AI, for AI, verified by humans