Kilobat with any MCP client

Install the CLI

The Kilobat CLI is free and open source, and it includes the MCP server. Install it with Homebrew:

Check it with kilobat --version. kilobat doctor says whether the Kilobat app is running, its network filter is on and Pro is active; the tools work without the app, which adds Wi-Fi names, history and rule enforcement. The CLI needs macOS 14 or later.

Connect the MCP server

1. Point your client at this command. It speaks MCP over stdin and stdout.

   kilobat mcp

2. Most clients read an mcpServers object like this one:

   {
     "mcpServers": {
       "kilobat": {
         "command": "kilobat",
         "args": [
           "mcp"
         ]
       }
     }
   }

3. Grok Build CLI, for example, adds it with one command:

   grok mcp add kilobat -- kilobat mcp

An app opened from the Dock doesn't always see the PATH your shell uses. If the server doesn't start, replace kilobat with the full path that command -v kilobat prints: /opt/homebrew/bin/kilobat on Apple silicon, /usr/local/bin/kilobat on Intel.

The server has 43 tools with full annotations (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) and returns the same JSON as the CLI. The docs list every tool.

Add the skills

Nine Agent Skills teach the agent to read Kilobat's measurements and to ask before changing settings. Install them with skills.sh, which asks which agents to install them for:

npx skills add seifip/kilobat-skills

For a client that skills.sh doesn't know, copy the folders from github.com/seifip/kilobat-skills to where it reads skills. Each skill is a folder with a SKILL.md.

The kilobat skill is the entry point: it holds the setup and safety rules and sends the agent to the skill that fits, such as mac-network-troubleshooting or mac-agent-runs. See all nine skills.

Try these prompts

Ask in plain words. The agent picks the tools; these are the ones each request usually needs.

What the agent can and can't change

Kilobat's MCP server marks every tool as read-only or as a change, and each change tool's description tells the agent to get your confirmation first. The kilobat skill holds the same rule: read freely, ask before changing.

Runs without asking

• Measure Wi-Fi, router, internet and DNS latency, and per-app traffic: diagnose_network, get_status, get_wifi, scan_wifi.
• Find the layer that stops one host, trace a route and compare resolvers: diagnose_connection, trace_route, benchmark_dns.
• Read history, sessions, rules and approvals: get_usage_report, list_sessions, list_rules, list_approvals.
• Ping for latency spikes with run_stutter_test, which sends pings only.

Asks you first

• A speed test, which uses up to 80 MB (run_speedtest; on a metered network, 10 MB or less), and a five-second LAN test (run_lan_test).
• Looking up your public IP with Cloudflare (get_exit_path with public_ip).
• Profiles, app rules, blocked hosts, the new-app mode, budgets, quotas, a network's data cost and pausing updaters (set_profile, set_app_rule, block_destination, set_new_app_mode, set_budget, set_app_quota, set_network_cost, set_updater_pause).
• DNS, which macOS also guards with an administrator password prompt (set_dns).
• Keeping the Mac running with the lid closed, every time (keep_awake with lid_closed). A plain keep_awake changes no settings and ends when it expires.

Can't do

• Let a waiting app connect on its own: decide_approval can block at once, but allowing opens Kilobat's approval window for you to confirm.
• Change AirDrop visibility: set_airdrop_visibility opens System Settings for you.
• Change sleep settings, another app's settings, or anything outside Kilobat. It never reads prompts or replies.
• Enforce rules without the Kilobat app and its network filter (Pro). Without them, rules are saved, not applied, and the agent should say so.

Keep your client's confirmation on for the change tools; never auto-approve them.

If the agent can also run shell commands, the kilobat CLI can do what the MCP tools do. The skills tell it which commands to run freely and which to ask about.

Track agent runs

An agent with its own hooks can tell Kilobat when a run starts, waits for you or is done by calling kilobat agent-event start|working|waiting|done|failed --agent <name> --session <id> --cwd <folder>. It prints nothing and always exits 0, so it can't break the agent; add --verbose when you try it by hand. Ready-made hooks exist for Claude Code, Codex, Gemini CLI and Cline. Without hooks, Kilobat goes by the agent's traffic and CPU.

Sources

• Model Context Protocol
• xAI docs: MCP servers in Grok Build
• Kilobat docs: AI agents and MCP
• skills.sh: supported agents and their skill folders

Checked on 30 October 2026. Agents change their settings often; if a step no longer matches, tell us at the support address.

Other agents