Skip to content

Compare

Ways to get AI teammates into Slack

There are three common ways to put AI teammates in Slack: run Squadmates in an AWS account in your organization, build your own Slack bot, or subscribe to a SaaS AI assistant for Slack. They differ in where your data goes, who builds the controls, and how access is scoped for each job.

At a glance

Three approaches.

A qualitative comparison. SaaS assistants differ a lot from vendor to vendor, so check each vendor's own documentation for where it runs and how it scopes access. For one of them compared row by row, with a source for every fact, see Squadmates vs Viktor.

Squadmates, building your own Slack bot, and SaaS AI assistants for Slack, compared
Squadmates in your AWS accountBuild your own Slack botSaaS AI assistant for Slack
Where it runsAn AWS account in your AWS Organization with a white-glove install (self-hosting coming soon), or an AWS account Agentifiers runs for youWherever you deploy itUsually the vendor's cloud
ControlsRouting gate, permission cards, audit log, pause, and versioned releases, includedYour team builds and maintains each oneThe vendor decides which controls exist
Identity per jobOne teammate per job, each with its own @handle and a named ownerWhatever you designVaries by vendor
When it actsOnly in conversations an owner or admin approved, and only when mentionedWhatever you designVaries by vendor
AuditEvery routing decision recorded, without message textYou build itVaries by vendor
Models and keysAmazon Bedrock in your account, or OpenAI or OpenRouter on your own keysYour choiceVaries by vendor
CodeSource available under FSL-1.1-ALv2. Source release coming soon.YoursVaries by vendor
Getting startedA white-glove install by AgentifiersYour team's timeOften a self-serve signup

Which fits

Which one fits.

Each approach is the right answer for someone. Squadmates has no self-serve signup today: installs are booked by email, managed access is invite-only, and self-hosting is coming soon.

  • Squadmates in your AWS account

    Fits when a security review matters, when several jobs each need their own owner and only the access they need, and when conversations should stay in your own AWS account. Read the security overview or what the install covers.

  • Building your own Slack bot

    Fits when you need one bot for one channel, have no audit needs, and have engineers to maintain it. See Squadmates vs building your own Slack bot.

  • A SaaS AI assistant for Slack

    Fits when you want to try something today with a self-serve signup, and a vendor holding your Slack conversations passes your review. Viktor is one example: see Squadmates vs Viktor.

Head to head

Two comparisons, in depth.

Each page compares Squadmates with one alternative, row by row, and says when the alternative is the better fit.

  • vs building your own Slack bot

    What a Slack AI bot needs before a whole company uses it: a routing gate, an identity and owner per bot, access policy, a secrets vault, an audit log, a pause switch, and versioned releases. Read Squadmates vs building your own Slack bot.

  • vs Viktor

    Viktor is one shared AI coworker for Slack and Microsoft Teams, with a self-serve signup and a long list of integrations. The page compares the identity and access model and where each one runs, says when Viktor is the better fit, and links every Viktor fact to Viktor's own pages. Read Squadmates vs Viktor.

Beyond Slack

More ways to implement AI agents.

Squadmates is one way to run AI agents. For a wider comparison of how to implement AI agents, read Agentifiers on ways to implement AI agents.

See it in your AWS account.

A white-glove install puts Squadmates and the Deployments.ai platform in an account in your AWS Organization.