MCP.so
ログイン

Omni Dev

@rust-works

Omni Dev について

概要はまだありません

基本情報

カテゴリ

その他

トランスポート

stdio

公開者

rust-works

投稿者

John Ky

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "omni-dev": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "ANTHROPIC_API_KEY",
        "-e",
        "ATLASSIAN_INSTANCE_URL",
        "-e",
        "ATLASSIAN_EMAIL",
        "-e",
        "ATLASSIAN_API_TOKEN",
        "-v",
        "${workspaceFolder}:/workspace",
        "-w",
        "/workspace",
        "ghcr.io/rust-works/omni-dev-mcp:latest"
      ],
      "env": {
        "ANTHROPIC_API_KEY": "<your-anthropic-api-key>",
        "ATLASSIAN_INSTANCE_URL": "https://<your-domain>.atlassian.net",
        "ATLASSIAN_EMAIL": "<your-email>",
        "ATLASSIAN_API_TOKEN": "<your-atlassian-api-token>"
      }
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Omni Dev?

Omni Dev is an intelligent Git commit message toolkit with AI-powered contextual intelligence. It transforms messy commit histories into professional, conventional commit formats using project-aware suggestions powered by Claude AI. Built with Rust for performance and safety, it also provides integrations with Atlassian (JIRA, Confluence), Datadog, YouTube transcripts, and Claude Code conversation export.

How to use Omni Dev?

Install Omni Dev with cargo install omni-dev or via Nix with nix profile install github:rust-works/omni-dev. Core commands include omni-dev git commit message twiddle to AI-improve commit messages, omni-dev git branch create pr to generate pull requests, omni-dev atlassian for JIRA and Confluence management, and omni-dev datadog for read-only Datadog queries. See the Getting Started guide for authentication and a 10-minute walkthrough.

Key features of Omni Dev

  • AI-powered commit message and PR description generation
  • Contextual awareness of project structure and conventions
  • Comprehensive analysis of commits, branches, and file changes
  • Safe amending of single or multiple commit messages
  • Atlassian JIRA and Confluence integration
  • Read-only Datadog metrics, monitors, logs, and SLOs querying
  • YouTube transcript fetching in multiple formats
  • Claude Code conversation history export to JSONL and Markdown
  • Conventional Commits automatic detection and formatting

Use cases of Omni Dev

  • Transform messy "fix stuff" or "wip" commits into conventional commit format with AI
  • Create professional pull requests with AI-generated descriptions, testing info, and details
  • Read, write, and manage JIRA issues and Confluence pages from the command line
  • Query Datadog metrics, monitors, logs, and SLOs without leaving the terminal
  • Export Claude Code chat history for behavioral analysis or work-log generation

FAQ from Omni Dev

What AI models does Omni Dev use?

Omni Dev uses Claude AI to analyze code changes and suggest meaningful commit messages and PR descriptions.

How is Omni Dev installed?

Omni Dev can be installed via cargo install omni-dev from crates.io or with Nix using nix profile install github:rust-works/omni-dev. A binary cache via Cachix is available for faster Nix builds.

Does Omni Dev support Atlassian products?

Yes, Omni Dev provides integrated commands for JIRA (read, write, edit, create, search, transition issues) and Confluence (read, search, create pages) with markdown and ADF JSON support.

What Datadog capabilities does Omni Dev offer?

Omni Dev provides read-only access to Datadog APIs including metrics queries, monitor lists, dashboard listings, log searches, and SLO queries. Authentication uses API key and App key, with support for different Datadog sites and on-prem overrides.

Can Omni Dev export Claude Code conversation history?

Yes, the omni-dev ai claude history sync command exports Claude Code chat sessions to .jsonl and/or .md files, grouped by project slug. It supports incremental sync, dry-run preview, pruning of removed sessions, and markdown rendering with structured tool call, user question, and error event formatting.

コメント

「その他」の他のコンテンツ