Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@omni-mcp-server search GitHub for MCP servers and save the top result to memory"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
One MCP server instead of six. GitHub, local Git, Slack, web fetching, persistent memory, and filesystem tools โ 15 tools in a single config entry, with dangerous capabilities off by default.
Setting up MCP usually means installing and configuring a separate server for every capability โ one for GitHub, one for filesystem, one for memory, one for fetch. Each has its own config block, its own runtime, its own startup cost.
Omni bundles the everyday developer tools into one server, one config entry, one npx command:
You'd normally installโฆ | With Omni |
| โ built in |
| โ built in |
| โ built in |
| โ built in |
a git server | โ built in |
a Slack server | โ built in |
๐ Safe by default โ shell execution and file writes are disabled unless you explicitly opt in to Agent Mode.
๐ Local & private โ runs on your machine over stdio. You bring your own tokens; nothing is proxied through a third party.
๐ชถ Zero install โ runs straight from npm via npx.
Related MCP server: Modular MCP Server
npx -y @bitofacoder/omni-mcp-server@latest setupThe wizard asks for your optional tokens (GitHub, Slack), asks whether to enable Agent Mode, and updates your Claude Desktop config for you. Restart Claude Desktop and you're done.
claude mcp add omni -e GITHUB_PERSONAL_ACCESS_TOKEN=your_gh_token -- npx -y @bitofacoder/omni-mcp-server@latestAdd this to your client's MCP config (e.g. claude_desktop_config.json):
{
"mcpServers": {
"omni": {
"command": "npx",
"args": ["-y", "@bitofacoder/omni-mcp-server@latest"],
"env": {
"GITHUB_PERSONAL_ACCESS_TOKEN": "your_gh_token",
"SLACK_BOT_TOKEN": "your_slack_token"
}
}
}
}All env vars are optional โ tools that need a missing token simply tell you so.
Group | Tool | What it does |
๐ GitHub |
| Search GitHub repositories |
| Get details of a specific issue | |
๐ณ Local Git |
| Working tree + staging area status |
| View local code changes | |
| View commit history | |
| Commit staged changes with a message | |
๐ป System |
| Read a local file |
| List a directory | |
| โ ๏ธ Write a file (Agent Mode only) | |
| โ ๏ธ Run a shell command (Agent Mode only) | |
๐ Web |
| Fetch any public URL as clean Markdown โ no API key needed |
๐ง Memory |
| Save a note/preference across chats |
| Retrieve a saved memory | |
| List everything remembered | |
๐ฌ Slack |
| Send a message to a Slack channel |
system_execute_command and system_write_file let the AI act on your machine โ run tests, install packages, edit files. Because that's real power, they are disabled by default and don't even appear in the tool list until you enable them:
"env": {
"OMNI_AGENT_MODE": "true"
}Everything else (reading files, git status, web fetch, memory) is read-only or sandboxed to its own data file.
PRs welcome โ especially new integrations (Linear, Notion, Jira, Discord) and Agent Mode hardening (allowlists, working-dir scoping). Open an issue first for bigger changes.
If Omni saved you some config wrangling, a โญ helps other people find it.
MIT โ see LICENSE.
Commit currently staged changes with a specific message.
Inputs: message (required).
Run git diff to see exact line-by-line code changes. Shows unstaged changes by default.
Inputs: staged.
Get details of a specific GitHub issue
Inputs: repo (required), owner (required), issue_number (required).
Search for GitHub repositories
Inputs: query (required).
Run git log to see recent commit history.
Inputs: maxCount.
Run git status to see the state of the working directory and staging area.
Inputs: no named parameters.
List all the memory keys currently saved in persistent storage.
Inputs: no named parameters.
Retrieve a previously saved memory by its key.
Inputs: key (required).
Save a note, preference, or piece of context to persistent memory so it can be recalled in future conversations.
Inputs: key (required), content (required).
Send a message to a Slack channel
Inputs: text (required), channel (required).
List contents of a local directory
Inputs: path (required).
Read a local file from the system
Inputs: path (required).
Fetch a public web page and read its content as clean Markdown
Inputs: url (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Scored across 13 tools
Each tool has a unique domain prefix (github_, slack_, system_, web_, memory_, git_) and a distinct action, making it clear which tool to use for a given task. There is no overlap in functionality between tools from different domains.
Most tools follow a <domain>_<verb> or <domain>_<verb>_<object> pattern (e.g., github_search_repos, slack_send_message), but git tools use nouns (git_status, git_diff, git_log) instead of verbs, breaking the consistency. Memory tools also use verbs without objects (memory_store) rather than verb_noun.
With 13 tools covering 6 different domains, the count is reasonable for an omnibus server. It provides a focused set of utilities without overwhelming the user, though some domains could benefit from a few more tools.
Each domain has significant gaps: GitHub only offers search and get issue (no create/update/delete), Slack only send message (no read/list), system only read and list dir (no write/delete), and git lacks add, push, pull, and branch operations. This can cause agent failures when users request common operations.
A MCP server built for developers enabling Git based project management with project and personalโฆ
An MCP server that provides read access to your cloud storage providers, bank accounts and more.
An MCP server that gives your AI access to the source code and docs of all public github repos
Nifty's MCP server โ exposes tasks, projects, messages, and files as tools for AI agents.
