switch

MCP server · GitHub · ★ 2.1k · graded 2026-10-29Source linked

Open-source AI teammates in Slack, Teams, Discord, Telegram and Mattermost. Self-host it, with any agent harness and any model.

No outside reports on file

Switch gives AI direct access to your team communications, so carefully limit what channels it can see, audit its permissions regularly, and only use it with models and platforms you trust.

Based on public adoption and transparency signals as of Oct 29, 2026. Our opinion, not a code audit or security guarantee.

Was this verdict for Switch helpful?

What it can access

  • Local filesNo local folder access notedLOW
  • Command executionNo shell command runner notedLOW
  • Network reachConnects to outside service APINOTED
  • CredentialsNo API token requiredLOW
  • Package sourceGitHub (sandbox-quantum/switch)PUBLIC

Inferred from this listing’s public package details and description. Check the project’s source before installing.

Install

No npm or PyPI package is listed. Install from source: github.com/sandbox-quantum/switch ↗

All setup options ↓

About Switch

Switch is an open-source tool that lets AI assistants work as team members in chat apps like Slack, Teams, Discord, Telegram, and Mattermost. You can run it on your own servers with any AI model and agent framework you choose.

Who it’s for

Teams that want AI assistants integrated into their existing chat platforms while keeping full control over the code and data. Good for organizations that prefer self-hosted solutions over cloud services.

Alternatives

  • Hugging Face Inference Endpoints with Slack integration (cloud-hosted AI).
  • OpenAI Assistants API (proprietary but widely supported).
  • Anthropic Claude for Slack (official but limited to Claude models).
  • N8N or Zapier workflows (general automation that can run AI tasks).

Setup

Check these steps against the project’s own instructions before running anything.

Follow the project's README at https://github.com/sandbox-quantum/switch (opens external site) for installation. The exact setup depends on your chat platform and hosting choice.

Why this grade

CC — Moderate risk signals

Signal-based grade (3/7 points): well starred on GitHub, public source code. This reflects adoption and transparency, not a code review.

Popularity and upkeep

36/100
GitHub stars
2.1k
Listing last refreshed
2026-10-29

These are public popularity signals, not a maintenance check. Look at the repository’s commit history and open issues.

Security reports

No reports found for this package in our research (last checked 2026-10-30).

Check a specific version →

Risks to watch for

  • Chat platform access: The tool can read and send messages in whatever channels or DMs you configure, potentially exposing sensitive conversations to the AI model.
  • Model data exposure: Whatever you chat about gets sent to your AI model (local or remote), so confidential information could leak depending on your setup.
  • Malicious prompts: Users in your chat can try to trick the AI into ignoring safety rules or revealing how it works.
  • Permission creep: If the AI has permissions to create channels, change settings, or access files, misconfiguration could cause unintended actions.
  • Supply chain risk: Third-party dependencies in the code could contain vulnerabilities; the project has 2037 GitHub stars but limited recent activity signals.

How to lower the risk

Least-privilege permissions

  • Grant the AI access only to specific channels, not your entire workspace.
  • Disable file uploads, user creation, or admin actions unless absolutely needed.

Network isolation

  • Run the model locally if handling sensitive data, not through a third-party API.
  • Use VPN or private networks between the bot and your chat platform.

Monitoring

  • Log all messages the AI reads and sends for audit trails.
  • Review AI responses periodically for unexpected behavior or data leaks.

Red flags to watch for

  • AI suddenly starts talking about confidential info it wasn't trained on.
  • Responses include instructions to bypass security or change settings.
  • Unusual message volume or requests to access new channels.

Questions

How do I install Switch?

Switch does not publish a standalone npm or PyPI package. Clone or follow the setup instructions in its source repository at github.com/sandbox-quantum/switch.

Are there security reports about Switch?

No reports found for this package in our research (last checked 2026-10-30). Switch currently holds Grade C (Moderate risk signals) based on public adoption and transparency signals. Key risk areas to watch include chat platform access, model data exposure, malicious prompts.

Is Switch actively maintained?

Switch has 2.1k GitHub stars. Our directory data was last refreshed on 2026-10-29. Check recent commit activity, release tags, and open issues at github.com/sandbox-quantum/switch before pinning it in production.

What are alternatives to Switch?

Notable alternatives to Switch include Hugging Face Inference Endpoints, OpenAI Assistants API, Anthropic Claude for Slack, N8N or Zapier workflows. You can also compare Switch side by side with other MCP servers in the Related tools section on this page.

README badge and team notes

Maintain or use Switch? Show its live grade in your GitHub README, docs site, or run a check from the terminal before you install.

AgentArmorHQ Risk Grade C for Switch

GitHub README Markdown:

markdown
[![AgentArmorHQ Risk Grade for Switch](https://agentarmorhq.com/api/badge/sandbox-quantum-switch)](https://agentarmorhq.com/skills/sandbox-quantum-switch)

HTML embed:

html
<a href="https://agentarmorhq.com/skills/sandbox-quantum-switch"><img src="https://agentarmorhq.com/api/badge/sandbox-quantum-switch" alt="AgentArmorHQ Risk Grade for Switch" height="20" /></a>

Terminal check:

terminal
curl -fsSL https://agentarmorhq.com/api/cli | node --input-type=module - check "sandbox-quantum-switch"

Share with your security lead

Need sign-off to use switch? Copy this note for Slack, a ticket or a pull request.

### AgentArmorHQ Security Review Memo: switch
- **Listing URL:** https://agentarmorhq.com/skills/sandbox-quantum-switch
- **Type:** App connector
- **Risk Grade:** Grade C (Moderate risk signals)
- **Outside Security Advisories on File:** 0
- **Public Adoption:** 2.1k GitHub stars

#### Permission & Blast-Radius Signals
- **Local files:** No local folder access noted
- **Command execution:** No shell command runner noted
- **Network reach:** Connects to outside service API
- **Credentials:** No API token required
- **Package source:** GitHub (sandbox-quantum/switch)

#### Recommended Hardened Setup
- **Source Repo:** https://github.com/sandbox-quantum/switch
- **Controls:** Pin an exact version in `.mcp.json`, scope access to a single workspace folder, and use a least-privilege read-only token where supported.

_Based on public adoption and transparency signals as of Oct 29, 2026. Our opinion, not a code audit or security guarantee._

Key facts

Type
App connector
Publisher
sandbox-quantum
Source code
github.com/sandbox-quantum/switch
Est. context load
~1.1k tokens / turn (~6 tool schemas, 0.53% of 200k context)
Popularity
2.1k GitHub stars
Last updated
2026-10-29
Last graded
2026-10-29

Community

Discussions for Switch

Ask questions or share configuration tips. Governed by our Community Rules & Terms.

No posts match this filter yet.

Be the first to ask a question or share a working tip.

Was this listing for Switch helpful?

Check before you install, in Claude

Your assistant answers Block, Review or No known issues for any package before it installs it.

Get this check inside Claude