KTME - Knowledge Tracking & Management Engine
Tracks and manages code changes from Git repositories, generates documentation automatically, and provides AI agents with intelligent access to service documentation through MCP server integration with feature mapping and search capabilities.
README
KTME - Knowledge Transfer Me
Automated documentation generation from Git changes using AI
KTME is a CLI tool and MCP server that automatically generates and maintains documentation from Git changes. It integrates with GitHub, GitLab, and Confluence, using AI to create meaningful documentation from code commits and pull requests.
Features
- Smart Documentation Generation - AI-powered documentation from Git diffs, commits, and PRs
- Multiple Integrations - GitHub, GitLab, and Confluence support
- Template System - Customizable Markdown templates with variable substitution
- MCP Server - Model Context Protocol server for AI agent integration
- Dual Storage - TOML and SQLite backends for flexibility
- Service Mapping - Organize documentation by service/project
Quick Start
Installation
# Install via npm (recommended - easiest)
npm install -g ktme-cli
# Install from crates.io
cargo install ktme
# Or build from source
cargo build --release
cargo install --path .
Basic Usage
# Generate docs from staged changes
ktme generate --service my-service --staged
# Extract GitHub PR and generate docs
ktme extract --pr 123 --provider github
ktme generate --service my-service --commit HEAD
# Update existing documentation
ktme update --service my-service --staged --section "API Changes"
# Map service to documentation location
ktme mapping add my-service --file docs/api.md
Configuration
Create ~/.config/ktme/config.toml:
[git]
github_token = "ghp_xxxxx"
gitlab_token = "glpat_xxxxx"
[confluence]
base_url = "https://your-company.atlassian.net/wiki"
api_token = "your-api-token"
space_key = "DOCS"
[ai]
provider = "openai"
api_key = "sk-xxxxx"
model = "gpt-4"
Documentation
- Quick Start Guide - Get started in 5 minutes
- Release Workflow - Publishing and version management
- Architecture - System design and components
- Development Guide - Contributing and development setup
Core Capabilities
1. Git Integration
Extract changes from various sources:
- Staged changes (
--staged) - Specific commits (
--commit abc123) - Commit ranges (
--range main..feature) - Pull/Merge requests (
--pr 123)
2. Documentation Generation
Generate documentation with templates:
# Use custom template
ktme generate --service api --template api-docs
# Generate changelog
ktme generate --service api --type changelog
# Output to specific file
ktme generate --service api --output docs/changelog.md
3. Smart Updates
Update existing documentation intelligently:
# Update specific section
ktme update --service api --section "Breaking Changes"
# Smart merge with existing content
ktme update --service api --staged
4. MCP Server
Run as MCP server for AI agents:
# Start server
ktme mcp start
# Available tools:
# - ktme_generate_documentation
# - ktme_update_documentation
# - ktme_list_services
# - ktme_search_features
Development
Quick Commands
# Test changes (fast - only new modules)
make test-changes
# Run all checks (format, lint, tests)
make pre-release
# Development cycle
make dev
Publishing
# Automated release workflow
make release
See docs/RELEASE.md for complete release documentation.
Architecture
┌─────────────┐ ┌──────────────┐ ┌───────────────┐
│ Git CLI │────▶│ Extractors │────▶│ Generators │
│ GitHub API │ │ (Diff/PR/MR) │ │ (Templates) │
│ GitLab API │ └──────────────┘ └───────────────┘
└─────────────┘ │ │
▼ ▼
┌──────────────┐ ┌───────────────┐
│ Storage │ │ Writers │
│ (TOML/SQLite)│ │ (MD/Confluence)│
└──────────────┘ └───────────────┘
Recent Updates (v0.1.0)
New Features
- ✅ Template engine with variable substitution
- ✅ Smart documentation merging by section
- ✅ GitHub PR extraction and integration
- ✅ GitLab MR extraction and integration
- ✅ Confluence writer with Markdown conversion
- ✅ Enhanced Markdown writer with section updates
Implementation Details
- 34 new tests (all passing)
- 10 files modified with new functionality
- Zero compilation errors after strict linting
See CHANGELOG.md for complete version history.
Contributing
We welcome contributions! Please see our Development Guide.
# Setup development environment
git clone https://github.com/FreePeak/ktme.git
cd ktme
make setup
# Run tests
make test-changes
# Submit PR
make pre-release
git push
License
MIT License - see LICENSE for details.
Support
- Issues: GitHub Issues
- Discussions: GitHub Discussions
- Documentation: docs/
Built with ❤️ using Rust
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。