subs MCP server
Commit 6 Nov 2026Sebuah agent harness yang menjalankan agent loop tanpa hak akses istimewa menggunakan MCP server untuk eksekusi tool, mendukung deployment lokal, remote, maupun tim.
Juga dikenal sebagai substructureai/subs
- Stars
- 13
- Tools
- —
- Unduhan, 30 hari
- —
- Versi terbaru
- —
- Commit terakhir
- 6 Nov 2026
- Lisensi
- MIT
Apa itu subs MCP?
Sebuah agent harness yang menjalankan agent loop tanpa hak akses istimewa menggunakan MCP server untuk eksekusi tool, mendukung deployment lokal, remote, maupun tim.
Ringkasan dibuat dari README proyek
Tentang subs MCP server
subs memungkinkan developer yang membangun AI agent berbasis cloud untuk menjalankan agent loop tanpa hak akses istimewa dengan kemampuan eksekusi tool. Harness ini mengelola seluruh lifecycle agent, termasuk durabilitas, retry, timeout, dan session state. Anda dapat melakukan deployment secara lokal melalui CLI, menjalankannya sebagai self-hosted server, atau menggunakannya melalui hosted engine. User mengonfigurasi agent menggunakan file toml, dengan mendefinisikan LLM provider seperti OpenRouter, Anthropic, atau OpenAI.
Developer dapat menghubungkan agent ke MCP server untuk tool spesifik, seperti Sentry atau Linear. Sistem ini mendukung berbagai mode deployment, termasuk mengintegrasikan agent secara langsung ke Slack melalui DM atau mention. Bagi Anda yang membutuhkan kontrol granular, harness ini memungkinkan agent diarahkan ke HTTP endpoint untuk menangani pengambilan keputusan melalui webhook. Setup ini memungkinkan logika kustom dalam bahasa pemrograman apa pun tanpa memerlukan SDK khusus, karena engine berkomunikasi melalui webhook standar dan JSON schema.
Apa yang bisa Anda lakukan dengannya
- 1Deploy agent ke SlackHubungkan agent ke workspace Slack agar dapat merespons direct messages atau mention di channel.
- 2Kontrol logika agent dengan webhookArahkan agent ke URL kustom untuk mencegat dan memodifikasi setiap keputusan yang diusulkan oleh engine.
- 3Kelola kredensial tool spesifik penggunaKonfigurasi agent untuk menggunakan koneksi berbasis user agar setiap orang berinteraksi menggunakan akun mereka sendiri.
- 4Eksekusi tugas background jangka panjangGunakan tool asinkron yang menerima panggilan dan melaporkan hasilnya kemudian, alih-alih memerlukan respons instan.
Ditulis berdasarkan README proyek. Setiap use case diperiksa terhadap kutipan dari README tersebut.
Konfigurasi
OPENROUTER_API_KEY
| Nama | Deskripsi | Tipe |
|---|---|---|
| OPENROUTER_API_KEY | — |
Pengecekan otomatis
Berjalan 15 Nov 2026Laporan lengkap →Belum diverifikasi. Sebuah server memerlukan skor 80, tidak ada temuan serius, dan data yang cukup untuk kami periksa. Cara kerja checks · Yang perlu diperbaiki untuk menambah poin
- Source repository terhubung dan dapat dijangkausubstructureai/subs is public and reachablegithub.com/substructureai/subs ↗Lolos
- Nama paket mencantumkan repositori ini sebagai sumbernyaTidak diperiksa
- Provenance build (atestasi Sigstore)Tidak diperiksa
- Lisensi open source (OSI)MITspdx.org/licenses/MIT.html ↗Lolos
- Ada commit atau release dalam satu tahun terakhirLast commit or release 2026-11-06github.com/substructureai/subs ↗Lolos
- OpenSSF ScorecardTidak diperiksa
- Nama package tidak mirip dengan package populerTidak diperiksa
- Tidak ada advisory kritis pada paket atau dependensinyaTidak diperiksa
- Tidak ada instruksi tersembunyi untuk model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersLolos
- Tools yang bersifat destruktif telah diberi labelTidak diperiksa
- Endpoint remote merespons handshake MCPTidak diperiksa
README
Lihat di GitHubsubs
Pre-1.0: APIs and the wire protocol can change between releases.
subs is an agent harness for the cloud.
It runs an unprivileged agent loop with no system access. It uses MCP servers for tools. It runs locally or as a client and a server.
Declare your agents in a config file. To customize the loop, point an agent at an HTTP endpoint and answer a webhook.
subs handles durability, retries, timeouts, MCP connection management, session state, session branching, AG-UI, Slack connection, LLM calls, subagents, interrupts and more.
To turn a sandbox into an MCP server, see mcpd.
Get started
Install the CLI
curl -fsSL https://subs.dev/cli.sh | bash
The CLI is also the engine.
Run an agent on your machine
Create a subs.toml.
name = "example"
[llm.openrouter]
type = "openrouter"
[agent.teammate]
llm = "openrouter"
model = "deepseek/deepseek-v4-flash-0731"
system = "You are a helpful teammate."
Set your provider key and talk to the agent.
export OPENROUTER_API_KEY=sk-or-...
subs chat teammate -c subs.toml
Run the engine as a server
Add a [serve] section and a [remote] that points at it.
[serve]
port = 9999
auth = false
[remote]
url = "http://localhost:9999"
Start the server.
subs serve -c subs.toml
In another terminal, the same chat command now talks to it.
subs chat teammate -c subs.toml
Run the agent on substructure.ai
Point [remote] at the hosted engine instead of your own.
[remote]
url = "https://api.substructure.ai"
Create the project from the file, then upload your LLM key.
subs apply
subs auth llm.openrouter
The same chat command now runs the turn on the hosted engine.
subs chat teammate -c subs.toml
Put the agent in Slack
Say which agent takes a DM and which one answers a mention.
[slack]
dm = "teammate"
mentions = "teammate"
Apply the file again, then connect your workspace.
subs apply
subs slack connect
Mention the bot in a channel and it answers in the thread.
Connect the agent to an MCP server
Declare the server and give it to an agent. Every user of the agent shares one credential.
[mcp.sentry]
url = "https://mcp.sentry.dev/mcp"
[agent.teammate]
llm = "openrouter"
model = "deepseek/deepseek-v4-flash-0731"
system = "You are a helpful teammate."
mcp = ["mcp.sentry"]
Authorize the connection.
subs auth mcp.sentry
Give each user their own MCP connection
Set credential = "user" and each user connects their own account. A
user-scoped connection works only in a one-on-one chat between the agent and
that user.
[mcp.linear]
url = "https://mcp.linear.app/mcp"
credential = "user"
[agent.personal]
llm = "openrouter"
model = "deepseek/deepseek-v4-flash-0731"
system = "Help me with my Linear issues."
mcp = ["mcp.linear"]
[slack]
dm = "personal"
mentions = "teammate"
Control the agent loop with a webhook
Point an agent at a URL and the engine sends every decision for that agent to your code. This gives you full control of the loop, including how the agent behaves in Slack.
[agent.teammate]
llm = "openrouter"
model = "deepseek/deepseek-v4-flash-0731"
system = "You are a helpful teammate."
mcp = ["mcp.sentry"]
worker = "https://example.com/agent"
Your endpoint reads the engine's proposal and returns it, changing only the steps you care about. There is no SDK to install.
import { serve } from "@hono/node-server";
import { Hono } from "hono";
import type { DecisionRequest, DecisionResponse } from "./protocol.ts";
function decide({ trigger, proposed }: DecisionRequest): DecisionResponse {
if (trigger.type === "session.start") {
return {
agent: {
...proposed.agent,
tools: [{ name: "current_time", description: "Get the current time" }]
}
};
}
// Run our tool when the model calls it.
if (trigger.type === "tool.execute" && trigger.name === "current_time") {
return { actions: [{ type: "tool.result", result: new Date().toISOString() }] };
}
// Accept the engine's proposal for everything else.
return proposed;
}
const app = new Hono();
app.post("/", async (c) => c.json(decide(await c.req.json())));
serve({ fetch: app.fetch, port: 4444 });
Only the agents that name a worker use one. The rest stay with the engine, in the same project and the same file.
Full walkthrough in the quick start. Docs: Workers, Connectors, Slack, Local development
Features
Slack
Mention the bot or DM it. The thread is the session. Route different channels to different agents.
Docs: Slack
Terminal chat
subs chat holds one session open, streams the reply as it is written, and
turns an approval prompt into a picker. The session is the same kind a Slack
thread is.
Docs: Chat
Examples: no-code-chat
Custom decisions
At each step the engine tells your code what it plans to do next. Accept the plan or do something else. A working agent is a few lines.
Docs: How it works
Any language
Your agent is an HTTP endpoint. Generate typed bindings from the published JSON schema.
Docs: Typed bindings
Examples: Go, Python, TypeScript, Elixir
MCP connectors
Declare an MCP server and the engine handles the authorization, reads the tools it offers, and runs every call. Your code never holds a token.
Docs: Connectors
Examples: Node
Plugins
Point an agent at an agent-plugins directory and it gets that plugin's skills and MCP servers.
Docs: Plugins
Any LLM
The engine calls Anthropic, OpenAI, or OpenRouter with your key. Or your worker makes the call and the engine never sees a key.
Docs: LLMs
Examples: Anthropic, OpenAI, OpenRouter
System One models
Your worker asks a fast model typed questions, such as TypeSafe's Jev or Cloudflare's Clef, and gets answers with a confidence. Route, guard, or score in the middle of a turn. The engine holds the key and saves every call.
Docs: System One models
Examples: Node
Crash recovery
Every step is saved before it runs. A run continues from where it stopped. The same message submitted twice runs once.
Docs: Durability
Human approval
An agent can stop and wait for a person to approve, then continue. A waiting agent uses no compute. In Slack this is a button.
Docs: Interrupts
Tools that take hours
A tool does not have to answer immediately. Accept the call, do the work on your own schedule, and report the result later.
Docs: Async tools
Chat history, editing, and branching
History, editing, regeneration, and branching belong to the engine. A user can edit an earlier message and go a new direction. The original branch stays.
Docs: Conversations
Existing chat UIs
The engine streams AG-UI events, so assistant-ui and CopilotKit connect to it directly.
Docs: AG-UI
Examples: assistant-ui, CopilotKit
Tools that run in the browser
A tool can run in the user's browser instead of on your server. The run waits for the browser, then continues.
Docs: Client-side tools
Examples: Node
Agent state without a database
The engine stores your agent's state with the conversation. Your code gets it on every request and writes changes back.
Docs: Agent state
Subagents
An agent can give work to other agents. Each child runs in its own session. The parent's totals include each child's cost and token use.
Docs: Subagents
Validated tool calls
Give a tool an input and output schema. The engine checks every call against it.
Docs: Tool calls
Retries and timeouts
Set a policy on any tool or model call. The engine applies it, and keeps applying it after a restart.
Docs: Retries and timeouts
Self-hosting
Run the engine on your own servers and hold every credential.
Docs: Self-hosting
Parts of the system
- Engine. Runs the agent loop, in Rust. It calls the model, runs tools, saves each step, retries failures, streams events, and supervises subagents. Use the hosted version at app.substructure.ai, run it from the CLI, or embed it in your process.
- Workers. Your agent code. It receives a trigger and returns actions. It runs in your codebase with your dependencies.
- Clients. They send work and stream events back, from your backend or from the browser. Slack and AG-UI are clients.
- CLI. Set up, deploy, watch, and debug from the terminal. It also runs the engine locally.
Install
curl -fsSL https://subs.dev/cli.sh | bash
The script verifies the release checksum and installs to ~/.local/bin. Set
SUBS_INSTALL_DIR to install elsewhere and SUBS_VERSION to pin a release.
Docs
Full documentation in docs/.
License
Pertanyaan tentang subs MCP
Bisakah saya menjalankan agent di Slack?
Ya, Anda bisa memasukkan agent ke dalam Slack dengan menentukan agent mana yang menangani DM dan mention di dalam file konfigurasi Anda.
Bagaimana subs menangani eksekusi tool?
subs menggunakan MCP server untuk tool dan dapat menangani otorisasi, pembacaan tool, serta eksekusi secara otomatis.
Bisakah saya menggunakan provider LLM saya sendiri?
Ya, engine ini dapat memanggil Anthropic, OpenAI, atau OpenRouter menggunakan API key yang Anda sediakan.