← 返回
未分类

Gitrama — Git History Intelligence

Git history intelligence powered by AI. Use when the user asks about git history, commit messages, branch names, repo insights, code archaeology, or wants to...
AI驱动的Git历史智能,用于查询Git历史、提交信息、分支名称、仓库洞察、代码考古或用户需求。
ahmaxdev
未分类 clawhub v1.0.1 1 版本 100000 Key: 无需
★ 1
Stars
📥 310
下载
💾 0
安装
1
版本
#latest

概述

Gitrama — Git History Intelligence

Gitrama is a CLI tool that brings AI-powered semantic search and intelligence to your Git history. It includes AskGIT, an embedded AI assistant that understands your entire repo context.

All AI processing is handled server-side via api.gitrama.ai. No API keys required — just install and use.

Capabilities

1. AskGIT — Semantic Git History Chat

Ask natural language questions about your repository:

  • "What changes were made to the authentication module last month?"
  • "Who has been the main contributor to the API layer?"
  • "Summarize the last 20 commits"
  • "What breaking changes happened between v1.0 and v2.0?"

Command: gtr chat

AskGIT automatically gathers context from 10 git subprocess calls (branch info, recent commits, diffs, contributors, file tree) and sends that context to api.gitrama.ai for intelligent analysis.

2. Smart Commit Message Generation

Generate conventional commit messages from staged changes:

  • Analyzes your git diff --staged
  • Produces properly formatted conventional commits (feat:, fix:, docs:, etc.)
  • Respects your project's commit conventions

Command: gtr commit

3. Branch Name Generation

Generate clean, descriptive branch names from a task description:

  • Follows git branch naming conventions
  • Includes type prefixes (feature/, bugfix/, hotfix/)
  • Keeps names concise and meaningful

Command: gtr branch

4. Git Log Analysis & Repo Insights

Get AI-powered analysis of your repository:

  • Contributor activity patterns
  • Code change frequency by area
  • Release history summaries
  • Technical debt indicators

Command: gtr chat then ask about repo insights

Installation

Gitrama is distributed via PyPI:

pip install gitrama

Verify installation:

gtr version

Configuration

No API keys or configuration needed. Gitrama connects to api.gitrama.ai for all AI processing out of the box.

Optional configuration available via:

gtr config

Usage Examples

When the user asks about git history or repo context, run the appropriate gitrama command:

# Start an interactive chat about the repo
gtr chat

# Generate a commit message for staged changes
gtr commit

# Generate a branch name
gtr branch "add user authentication with OAuth"

# Quick shortcuts
gtr c   # alias for commit
gtr b   # alias for branch
gtr ch  # alias for chat

Data Flow

Local CLI (gtr) → gathers git context locally → POST to api.gitrama.ai → AI response
  • Git context (commits, diffs, branch info) is collected locally via git subprocess calls
  • Context is sent to api.gitrama.ai for AI processing
  • No API keys are stored or required on the user's machine
  • No sensitive IP runs locally — the intelligence layer lives server-side

Important Notes

  • Gitrama must be run inside a git repository
  • AskGIT maintains a 40-message conversation history for multi-turn context
  • The tool collects repo context automatically — no manual configuration needed
  • Repository data sent to api.gitrama.ai is used only for generating responses and is not stored

Error Handling

If gtr is not found after install:

# Ensure pip scripts are on PATH
python3 -m gitrama --version

If connection errors occur:

# Verify api.gitrama.ai is reachable
curl -s https://api.gitrama.ai/health

Links

  • Website: https://gitrama.ai
  • PyPI: https://pypi.org/project/gitrama/

版本历史

共 1 个版本

  • v1.0.1 当前
    2026-05-12 05:39 安全 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

security-compliance

Skill Vetter

spclaudehome
AI智能体技能安全预审工具。安装ClawdHub、GitHub等来源技能前,检查风险信号、权限范围及可疑模式。
★ 1,223 📥 267,358
ai-intelligence

Self-Improving + Proactive Agent

ivangdavila
自我反思+自我批评+自我学习+自组织记忆。智能体评估自身工作、发现错误并持续改进。
★ 1,371 📥 319,705
ai-intelligence

self-improving agent

pskoett
捕获经验教训、错误及修正内容,以实现持续改进。适用于以下场景:(1)命令或操作意外失败;(2)用户纠正Claude(如“不,那不对……”“实际上……”);(3)用户请求的功能不存在;(4)外部API或工具出现故障;(5)Claude发现自身
★ 4,072 📥 805,155