MCP server omni-dev
Коммит 10 нояб. 2026 г.Инструментарий на базе ИИ для переписывания git-коммитов, генерации pull requests и интеграции через MCP с Jira, Confluence и Datadog.
Также известен как rust-works/omni-dev
- Звёзды
- 2
- Инструменты
- —
- Загрузки, 30 дней
- —
- Последняя версия
- —
- Последний коммит
- 10 нояб. 2026 г.
- Лицензия
- BSD-3-Clause
Что такое omni-dev MCP?
Инструментарий на базе ИИ для переписывания git-коммитов, генерации pull requests и интеграции через MCP с Jira, Confluence и Datadog.
Сводка сгенерирована из README проекта
О omni-dev MCP server
Разработчики, стремящиеся навести порядок в истории версионирования, могут использовать omni-dev для преобразования хаотичных сообщений коммитов в профессиональные форматы conventional commits. Используя Claude AI для анализа изменений в коде, инструмент предлагает осмысленные описания, учитывающие структуру проекта и рабочие процессы. Он написан на Rust для обеспечения высокой производительности и включает функции безопасности, такие как проверка рабочей директории, чтобы предотвратить случайные изменения в удаленных ветках.
Помимо управления Git, этот инструментарий предоставляет интеграцию MCP для работы с Jira, Confluence и Datadog. Это позволяет пользователям управлять задачами в agile, просматривать логи рабочего времени, а также работать с историей страниц или вложениями в Confluence через ИИ-клиент. Система также включает браузерный мост для выполнения HTTP-запросов через аутентифицированную вкладку браузера, что помогает работать с дашбордами за SSO, такими как Grafana или Loki, не раскрывая конфиденциальные токены.
Пользователи могут установить инструмент через cargo или Nix. Он поддерживает несколько бэкендов ИИ, включая Anthropic, Bedrock, OpenAI и Ollama. Сервер предназначен для локального использования с управляемым демоном, который ведет реестр worktrees и лог запросов.
Что вы можете с ним делать
- 1Улучшение небрежной истории коммитовИИ-клиент может переписать диапазон неорганизованных коммитов в профессиональный формат conventional commits, используя контекстуальный интеллект.
- 2Генерация описаний для pull requestsВы можете дать команду ИИ создать подробный pull request, включающий развернутые описания и информацию о тестировании.
- 3Доступ к дашбордам под защитой SSOИспользуйте браузерный мост для выполнения запросов через аутентифицированную вкладку браузера для таких сервисов, как Grafana.
Взято из README проекта. Каждый вариант использования проверен на соответствие цитате из него.
Конфигурация
OMNI_BRIDGE_TOKENOMNI_DEV_GATEWAY_UPSTREAM_URLRUST_LOG
| Название | Описание | Тип |
|---|---|---|
| OMNI_BRIDGE_TOKEN | — | |
| OMNI_DEV_GATEWAY_UPSTREAM_URL | — | |
| RUST_LOG | — |
Автоматические проверки
Запущено 14 нояб. 2026 г.Полный отчёт →Еще не проверено. Для сервера требуется оценка 80, отсутствие серьезных проблем и достаточно данных для нашей проверки. Как работают проверки · Что исправить, чтобы получить больше баллов
- Связанный и доступный исходный репозиторийrust-works/omni-dev is public and reachablegithub.com/rust-works/omni-dev ↗Пройдено
- Пакет указывает этот репозиторий как источникНе проверялось
- Происхождение сборки (аттестация Sigstore)Не проверялось
- Open source лицензия (OSI)BSD-3-Clausespdx.org/licenses/BSD-3-Clause.html ↗Пройдено
- Коммит или релиз за последний годLast commit or release 2026-11-10github.com/rust-works/omni-dev ↗Пройдено
- OpenSSF ScorecardНе проверялось
- Имя пакета не похоже на популярный пакетНе проверялось
- Нет критических предупреждений в пакете или его зависимостяхНе проверялось
- Нет скрытых инструкций для модели0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersПройдено
- Деструктивные инструменты помеченыНе проверялось
- Удалённая конечная точка отвечает на рукопожатие MCPНе проверялось
README
Открыть на GitHubomni-dev
An intelligent Git commit message toolkit with AI-powered contextual intelligence. Transform messy commit histories into professional, conventional commit formats with project-aware suggestions.
🎬 See It In Action
Watch omni-dev transform messy commits into professional ones with AI-powered analysis
30-Second Demo
Transform your commit messages and create professional PRs with AI intelligence:
# Analyze and improve commit messages in your current branch
omni-dev git commit message twiddle 'origin/main..HEAD' --use-context
# Before: "fix stuff", "wip", "update files"
# After: "feat(auth): implement OAuth2 authentication system"
# "docs(api): add comprehensive endpoint documentation"
# "fix(ui): resolve mobile responsive layout issues"
# Create a professional PR with AI-generated description
omni-dev git branch create pr
# 🎉 Generates comprehensive PR with detailed description, testing info, and more
✨ Key Features
- 🤖 AI-Powered Intelligence: Claude AI analyzes your code changes to suggest meaningful commit messages and PR descriptions
- 🧠 Contextual Awareness: Understands your project structure, conventions, and work patterns
- 🔍 Comprehensive Analysis: Deep analysis of commits, branches, and file changes
- ✏️ Smart Amendments: Safely improve single or multiple commit messages
- 🚀 PR Creation: Generate professional pull requests with AI-powered descriptions
- 📦 Automatic Batching: Handles large commit ranges intelligently
- 🎯 Conventional Commits: Automatic detection and formatting
- 🌐 Browser Bridge: Drive HTTP requests through an authenticated browser tab without exfiltrating cookies or tokens
- 🗂️ Worktrees View: One live view of every repo and git worktree open across all your VS Code windows
- 🛡️ Safety First: Working directory validation, protection against amending commits already in remote main branches, and error recovery
- ⚡ Fast & Reliable: Built with Rust for memory safety and performance
🚀 Quick Start
Installation
# Install from crates.io
cargo install omni-dev
# Install with Nix
nix profile install github:rust-works/omni-dev
# Install with Nix flakes (development)
nix run github:rust-works/omni-dev
Pre-built binaries are attached to each
release. The Linux ones
(omni-dev-linux.tar.gz, omni-dev-linux-arm64.tar.gz) are dynamically linked
and need glibc 2.35 or newer (Ubuntu 22.04 and Debian 12 qualify). On an older
host, such as RHEL 9 or Amazon Linux 2023 (glibc 2.34), they fail in the loader
with version `GLIBC_2.xx' not found; build from source with
cargo install omni-dev instead. The release workflow fails if a binary needs
more than that floor (docs/RELEASE.md).
Next step: see Getting Started — a 10-minute walkthrough from authentication to your first AI-improved commit. (For just the API-key reference, see Authentication.)
Shell Completion
omni-dev completions <shell> prints a completion script to stdout for
bash, zsh, fish, powershell, or elvish. The quickest path is bash
per-user:
# Add to ~/.bashrc:
eval "$(omni-dev completions bash)"
See docs/shell-completion.md for per-shell install
recipes, the $fpath/compinit setup zsh requires, and troubleshooting.
🆚 How omni-dev Compares
omni-dev sits in two adjacent spaces — AI commit-message tooling and
Atlassian/dev-workflow MCP servers. The tables below contrast the
incumbents on the dimensions a first-time reader is most likely to weigh.
In every cell, ✅ means full / native support, ⚠ means partial or
available only with caveats, and ❌ means not supported — and omni-dev's
own limitations are flagged just as honestly (the ⚠ marks in its own
columns).
Beyond these two niches, omni-dev also ships a supervised daemon that
hosts a browser bridge (an authenticated proxy that runs requests
through a logged-in browser tab for SSO-gated dashboards such as Grafana
and Loki), a Snowflake SQL service (one external-browser SSO session
reused for concurrent queries), and a worktrees registry (one live view of
the repos open across every VS Code window), plus a local append-only
request log (omni-dev log). These have no direct incumbent in either
table below, so
they are called out here rather than scored against tools that don't aim
for them.
vs AI commit tools
| omni-dev | opencommit | aicommits | |
|---|---|---|---|
| Rewrite existing commits in a range | ✅ twiddle |
❌ pre-commit only | ❌ pre-commit only |
| Parallel batched processing (long ranges) | ✅ --concurrency N |
❌ | ❌ |
| AI-written PR descriptions | ✅ git branch create pr |
⚠ GitHub Action only | ❌ |
| Project-context awareness | ✅ --use-context |
❌ | ❌ |
Sandboxed claude-cli backend |
✅ ADR-0028 | ❌ | ❌ |
| Multi-backend (Anthropic / Bedrock / OpenAI / Ollama) | ✅ | ✅ | ✅ |
| Conventional Commits | ✅ | ✅ | ⚠ config |
| Language / runtime | Rust (static binary) | Node.js | Node.js |
vs Atlassian-workflow MCP servers
omni-dev's MCP server also exposes Git tools (commit analysis, twiddling,
PR creation), Datadog tools, and an ai_chat proxy — surfaces the
Atlassian-focused servers don't aim for. The table below compares only
Atlassian capability depth.
| omni-dev MCP | sooperset/mcp-atlassian | Atlassian official (Rovo) | |
|---|---|---|---|
| Jira REST surface | ✅ 36 tools (agile, fields, dev panel, links, watchers, worklogs, versions, changelog) | ✅ 49 tools (above + JSM, proforma forms, SLA, batch ops) | ⚠ 14 tools (basic CRUD, search, transitions, worklogs only) |
| Confluence REST surface | ✅ 25 tools (history, diff, attachments, labels, spaces, inline + footer comments) | ✅ 24 tools (history, diff, attachments, labels; no inline comments / spaces) | ⚠ 12 tools (inline + footer comments, spaces; no delete / move / history / diff / attachments / labels) |
| Lossless JFM ↔ ADF round-trip | ✅ full ADF node set (schema v56.1.18) + unsupported-node escape | ❌ | ⚠ raw ADF, model-dependent |
| Anchored review-comment preservation | ✅ annotation marks survive round-trip | ❌ anchor stripped, comments orphaned | ⚠ ADF carries anchors; model-dependent |
| Pre-flight ADF schema validation | ✅ nesting + arity, before write | ❌ | ❌ |
| Offline JFM ↔ ADF conversion (no creds) | ✅ atlassian_convert |
❌ | ❌ |
| Cloud + Server + Data Center | ⚠ Cloud verified | ✅ Cloud + Server (v6+) + DC (Jira v8.14+) | ❌ Cloud only |
| Auth | ⚠ API token only | ✅ API token / PAT / OAuth 2.0 | ✅ OAuth 2.1 / API token |
Last verified: 2026-06-23. omni-dev and sooperset rows are live-tested — a
tools/list enumeration (omni-dev branch build vs
ghcr.io/sooperset/mcp-atlassian:latest) plus a live read→write→read fidelity
cycle on a complex page. Atlassian Rovo's server accepts the API token but
gates tool execution behind an org-admin grant, so its rows combine
Atlassian's
Supported tools
docs with the ADF-passthrough reasoning (raw ADF can round-trip, but only if
the model echoes it faithfully — no deterministic guarantee), not a live run.
Refresh quarterly or whenever a release-note search for the comparators flags
a relevant change.
📋 Core Commands
🤖 AI-Powered Commit Improvement (twiddle)
The star feature - intelligently improve your commit messages with real-time model information display:
# Improve commits with contextual intelligence
omni-dev git commit message twiddle 'origin/main..HEAD' --use-context
# Process large commit ranges with parallel processing
omni-dev git commit message twiddle 'HEAD~20..HEAD' --concurrency 5
# Save suggestions to file for review
omni-dev git commit message twiddle 'HEAD~5..HEAD' \
--save-only suggestions.yaml
# Auto-apply improvements without confirmation
omni-dev git commit message twiddle 'HEAD~3..HEAD' --auto-apply
🔍 Analysis Commands
# Analyze commits in detail (YAML output)
omni-dev git commit message view 'HEAD~3..HEAD'
# Analyze current branch vs main
omni-dev git branch info main
# Get comprehensive help
omni-dev help-all
🚀 AI-Powered PR Creation
Create professional pull requests with AI-generated descriptions:
# Generate and create PR with AI-powered description
omni-dev git branch create pr
# Create PR with specific base branch
omni-dev git branch create pr main
# Save PR details to file without creating
omni-dev git branch create pr --save-only pr-description.yaml
# Auto-create without confirmation
omni-dev git branch create pr --auto-apply
📝 Atlassian Integration
Read, write, and manage JIRA issues and Confluence pages from the command line:
# Authenticate with Atlassian Cloud
omni-dev atlassian auth login
# Check authentication status
omni-dev atlassian auth status
# Fetch a JIRA issue as markdown
omni-dev atlassian jira read PROJ-123
# Fetch as raw ADF JSON
omni-dev atlassian jira read PROJ-123 --format adf
# Push markdown changes back to JIRA
omni-dev atlassian jira write PROJ-123 issue.md
# Interactive edit: fetch, edit in $EDITOR, push
omni-dev atlassian jira edit PROJ-123
# Search issues with JQL
omni-dev atlassian jira search --project PROJ --status Open
# Create an issue
omni-dev atlassian jira create issue.md --project PROJ --summary "Fix bug"
# Transition an issue
omni-dev atlassian jira transition PROJ-123 "In Progress"
# Confluence: read, search, create pages
omni-dev atlassian confluence read 12345
omni-dev atlassian confluence search --space ENG --title auth
omni-dev atlassian confluence create page.md --space ENG --title "New Page"
# Convert markdown to ADF JSON (offline)
omni-dev atlassian convert to-adf input.md
Self-hosted PATs use atlassian auth login --auth-mode bearer without email.
See PAT authentication and compatibility limits;
most service operations still target Cloud APIs.
📊 Datadog Integration (read-only)
Authenticate against the Datadog API and query metrics, monitors, dashboards, logs, events, SLOs, hosts, and downtimes. See the Datadog integration guide for the full subcommand reference, authentication setup, rate-limit behaviour, and troubleshooting.
# Configure Datadog API credentials (prompts for API key, APP key, and site)
omni-dev datadog auth login
# Verify the credentials by calling /api/v1/validate
omni-dev datadog auth status
# Query metrics, monitors, dashboards, logs, and SLOs
omni-dev datadog metrics query --query 'avg:system.cpu.user{*}' --from 15m
omni-dev datadog monitor list --tags env:prod
omni-dev datadog dashboard list
omni-dev datadog logs search --filter 'service:api status:error' --from 1h
omni-dev datadog slo list --tags team:platform
DATADOG_SITE defaults to datadoghq.com. Other regions (datadoghq.eu,
us3.datadoghq.com, us5.datadoghq.com, ap1.datadoghq.com, ddog-gov.com)
are recognised without warning. Environment variables DATADOG_API_KEY,
DATADOG_APP_KEY, DATADOG_SITE override the stored settings. For on-prem
or proxied installs, set DATADOG_API_URL to override the site-derived URL.
All Datadog subcommands are also exposed as MCP tools (datadog_*) — see
docs/mcp.md. For the full guide covering
every family with worked examples, see docs/datadog.md.
📧 Gmail and 📁 Drive (moved to gwi)
These commands, their MCP tools and library modules now live in
gwi (Google Workspace Interface), a standalone
project (ADR-0095). gwi import copies your existing accounts and
settings from ~/.omni-dev; see the user guide.
For one release, omni-dev gmail ... and omni-dev drive ... remain as hidden stubs that
accept any arguments and exit non-zero with a message pointing at gwi; the next minor
release deletes them.
🎙️ Transcript Fetching
Pull captions and transcripts from external media platforms. YouTube is the first supported source; the CLI namespace and library are designed so additional sources (Vimeo, podcast RSS, generic VTT/SRT URLs) can be added without restructuring. See docs/transcript.md for the full reference and the recipe for adding a new source.
# Fetch captions for a YouTube video as SubRip (default).
omni-dev transcript youtube fetch https://www.youtube.com/watch?v=jNQXAC9IVRw
# WebVTT to a file, falling through to auto-generated captions if needed.
omni-dev transcript youtube fetch jNQXAC9IVRw \
--format vtt --auto --output me-at-the-zoo.vtt
# Synthesise a translated track when no native French track exists.
omni-dev transcript youtube fetch <url> --lang fr --translate fr
# List available caption tracks (manual + auto-generated).
omni-dev transcript youtube list-langs <url>
# Show video metadata (title, channel, duration, languages).
omni-dev transcript youtube info <url> --output json
--format accepts srt, vtt, txt, or json. Locators may be a
watch?v= URL, a youtu.be/ short URL, a /shorts/ or /embed/ URL,
or a bare 11-character video ID. Age-gated and login-required videos
surface as a typed PlayabilityRefused error carrying YouTube's status
code rather than a generic HTTP failure.
🌐 Browser Bridge
Drive HTTP requests through an authenticated browser tab. When you are investigating internal services (Grafana/Loki, internal dashboards, SSO-gated admin panels), the browser already holds sessions — SSO, OAuth, cookies — that are hard to replicate programmatically. The bridge issues requests inside the browser's authenticated context without exfiltrating cookies or tokens (a confused deputy by design). Both planes are authenticated and default-closed; see docs/browser-bridge.md for the full guide and ADR-0036 for the security rationale.
# Start the bridge; it prints the bound ports, a session token, and a JS
# snippet to paste into the DevTools console of the authenticated tab.
omni-dev browser bridge serve
# Drive requests through the tab (token from the bridge's stdout).
export OMNI_BRIDGE_TOKEN=<token printed by the bridge>
omni-dev browser bridge request --url /loki/api/v1/labels
# POST a JSON payload from a file, with a custom header.
omni-dev browser bridge request --url /api/foo --method POST \
--body @payload.json --header "Accept: application/json"
# Stream a long-lived endpoint (SSE / chunked) instead of buffering.
omni-dev browser bridge request --url /api/events --stream
# Route to a specific tab when several are connected (by id or origin).
omni-dev browser bridge request --url /api/foo --target https://grafana.internal
Supports binary and streaming response bodies, multi-tab routing via
X-Omni-Bridge-Target, per-request --credentials and --allow-origin
overrides, and a transparent proxy for tools that speak plain HTTP.
🛰️ Daemon
Host long-lived services in one supervised process behind a private per-user
Unix-domain control socket. The browser bridge is the first service migrated
onto it (Snowflake and the worktrees registry followed), and on macOS an
optional menu-bar app gives live control. daemon start installs a launchd LaunchAgent for auto-start at
login, and status reports every hosted service. See
Running under the daemon and
ADR-0039 for the architecture.
# Start the background daemon (installs a launchd LaunchAgent on macOS)
omni-dev daemon start
# Per-service status (add --json for machines)
omni-dev daemon status
# Restart or stop it
omni-dev daemon restart
omni-dev daemon stop
The daemon is Unix-only — its control plane is a Unix-domain socket — while the rest of omni-dev runs everywhere.
❄️ Snowflake
Authenticate a Snowflake session once via external-browser SSO, then run concurrent arbitrary SQL across any account without an SSO popup on every query. The daemon holds the session in memory and multiplexes a bounded pool, so each query can still set its own warehouse/role/database/schema. See docs/snowflake-service.md.
# Run SQL (from an argument or stdin); the first query opens the SSO browser
omni-dev snowflake query "select current_version()"
# Per-query context overrides and JSON output
omni-dev snowflake query "select * from t limit 10" \
--warehouse WH --role ANALYST --database DB --schema PUBLIC --format json
# Inspect or evict live sessions
omni-dev snowflake sessions
omni-dev snowflake disconnect --account <ACCOUNT> --user <USER>
Account/user/context default from SNOWFLAKE_* env vars then
~/.omni-dev/settings.json — no accounts are hardcoded. Runs on the daemon, so
it is Unix-only.
📓 Request Log
Every invocation and the HTTP requests it issues are recorded to a local, append-only log you can search and tail. Best-effort and default-on; no secret is ever written (auth headers are redacted, bodies opt-in). See docs/log.md.
# Recent activity (one line each)
omni-dev log
# Filter by service and status class, or a query expression; follow live
omni-dev log --service jira --status 5xx
omni-dev log --query 'method:POST AND status:4xx' --follow
# Full records as JSON (byte-identical to the on-disk lines)
omni-dev log --format json -n 20
Set OMNI_DEV_LOG_DISABLE=1 to turn it off, or OMNI_DEV_LOG_BODIES=1 /
OMNI_DEV_LOG_HEADERS=1 to opt into capturing bodies/headers.
🗂️ Worktrees
See every repo and git worktree open across all your VS Code windows in one live view. A VS Code extension host is sandboxed per window — no extension alone can see a sibling window's folders — so a small first-party companion extension registers each window with the daemon, which aggregates them into a single registry served back to the CLI, tray, and extension UI. The registry is in-memory only; windows that crash without unregistering age out automatically. See docs/worktrees-service.md and ADR-0040.
# One line per open window and its folders (add --json for machines)
omni-dev worktrees list
Runs on the daemon, so it is Unix-only.
🤖 Agent Sessions
Track the Claude Code, Codex and pi.dev sessions running across every terminal
and VS Code window, each with a coarse live state (working, idle, or waiting on
you). Opt-in hooks, the claude-wrap/codex-wrap wrappers, a pi.dev extension
and transcript watchers feed the daemon, which keeps the sessions in memory and
serves them to the CLI, the tray, worktrees ui and the VS Code worktrees view.
See docs/sessions-service.md.
# Install the Claude Code (and, when present, Codex and pi.dev) hooks
omni-dev sessions install-hooks
# One line per live session
omni-dev sessions list
Runs on the daemon, so it is Unix-only.
Вопросы о omni-dev MCP
Что omni-dev может делать с git-коммитами?
С помощью контекстуального интеллекта на базе ИИ инструмент может анализировать и преобразовывать небрежную историю коммитов в профессиональный формат conventional commits.
Поддерживает ли он генерацию pull requests?
Да, он может создавать профессиональные pull requests с описаниями на базе ИИ, включая информацию о тестировании.
