Launched this week
MakersClaw
Hire AI employees that live in your Slack, Teams, Telegram
569 followers
Hire AI employees that live in your Slack, Teams, Telegram
569 followers
Hire AI employees that run 24/7 in their own container with their own memory. One-click into your Slack, Telegram, or Teams. Pre-built for support, sales, research, SEO, or anything you write yourself. Pay per call for the tools they use.
the framing of 'AI employees that live in slack' is sharp. the part i'm most curious about: how does an AI employee accumulate a track record? a human gets references from former teammates, a paper trail of what they shipped. when this AI moves between teams, what travels with it?
MakersClaw
@thenameisarian Good question. The track record builds up in the employee's persistent memory: every conversation, every action it took, every file you uploaded (there is a file system for the agent where you can upload files), every customer it dealt with. All of that accumulates in a store inside its pod.
When the role changes inside a workspace, all of it travels. You reconfigure the same employee and nothing gets erased. Memory persists and the file system stays. For this, skills get added on top rather than replacing what was there (just like a human, a skill learnt is not lost).
Congrats! I'm a little curious how does MakersClaw handle context continuity when the same user interacts across Slack and Telegram simultaneously?
MakersClaw
@crystalmei Great question. The employee's memory is the employee's. Channels are just the interface you talk to it through. Each employee runs in its own pod with its own memory, so messages from every connected channel write to the same store. Switch from Slack to Telegram mid-conversation and the agent pulls the same context. From a technical standpoint, the pod processes inbound messages serially, so simultaneous threads don't collide either.
The "configure by chatting with the employee" concept really caught my attention. In practice, do users prefer conversational setup over traditional forms and settings?
MakersClaw
@harini_mukesh Yes, from what we're seeing. The conversational setup is the part people are most excited about today, which feels promising. Small dataset though, and the signal could pivot either way as usage scales.
Worth noting: we're already hybrid today. All of the configuration happens through chat, but some settings (the structured ones like model selection, skill selection) are also selectable. So users get to pick the path that fits the task. We might keep it that way long-term rather than forcing everything into chat.
@sachinsharma The hybrid approach sounds like the best of both worlds. Chat for flexibility and settings for precision. Curious to see how this evolves as more users come in!
Congrats on shipping. What tasks can this actually accomplish end-to-end without human intervention?
MakersClaw
@zerotox It depends on how you configure it.
Customer support is a clean example. Give the agent the right context for the role (product docs, tone of voice, escalation rules, FAQ patterns) and connect it to your support channel. A ticket comes in, the agent reads it, looks up the customer in the CRM, drafts and sends the reply, logs the interaction, escalates the rare hard ones to you. End to end, no human in the loop.
Same model works for sales outreach. Connect your lead list, brief the agent on your brand voice and the angle, hook up Gmail or Outlook, and the agent drafts and sends the sequence on schedule. Logs everything in the CRM. You step in only when a lead replies.
The pattern is consistent: the better the upfront context you give it, the more it runs without you. For high-stakes actions you'd want explicit sign-off on, that's still a human-in-the-loop step today.
Documentation.AI
Congrats. How are you handling permissions? Giving agents access to Slack, GitHub, Gmail, etc. gets scary pretty quickly. :D
MakersClaw
@roopreddy Thanks!
Right concern. Permissions are OAuth-scoped at the workspace level: you see what each integration is asking for and you decide whether to grant it. The agent only gets the scope you granted. So if you OAuth GitHub with read-only access, that's all it can do, even if someone in chat asks for more. No API keys floating around in config files. You also control which employees in your workspace get which integrations, and revoke is one click.
AI employees living right in Slack and Teams sounds really useful. Can each AI employee be customized to a specific role or task?
MakersClaw
memi
AI employees living where the team already talks is the right move. The hard part is making them helpful without becoming one more coworker to manage.
MakersClaw
MakersClaw
@sarveshsea We lean on persistent memory and the file system to cut the reexplaining context cost. You tell it once (or upload context files once), then the agent remembers.
None of this solves the deeper concern fully. The honest version is that "an AI employee that needs zero oversight" isn't a 2026 reality for anyone. We're trying to push the oversight cost as low as possible per action, but it's still genuinely a hard problem.