Agensis turns AI agents into coworkers in a shared Slack-like workspace
jasonkneen · x · 2026-07-28
agensis is positioning itself as a shared workspace where humans and AI agents work together in channels and threads, with persistent memory, visible presence, and a shared canvas.
The product pitch emphasizes that it is not a traditional agent orchestrator. Instead of defining agents in isolation and sending them off to separate branches, agensis lets teams invite agents into the workspace like colleagues. The key workflow pieces are:
- agents live in channels and DMs alongside people
- mentions in threads hand work to an agent with full context
- threads can be forked into parallel branches and merged back later
- memory persists across sessions, so agents retain preferences, facts, and files
- agents can delegate to each other, with live visibility into handoffs and heartbeats
The screenshot shows a Slack-like interface with channels, threads, agents, and a profile panel for an agent named amp, suggesting the product is trying to make agent collaboration feel like a normal team chat experience rather than a separate automation console.
More from coding & agent
- Meetup will build a Claude Code-style coding agent from scratch, no framework — dfinke · 2026-07-28
- OpenAI’s Codex is called open source, but the UI code still seems missing — lucasmeijer · 2026-07-28
- A recursive agent joke captures how absurd multi-agent workflows can get — gethackteam · 2026-07-28
- Tutorial shows how to publish Instagram posts from Claude with a Contentdrips MCP server — pubgupdates · 2026-07-28
- Stream pitches a dating-app stack that hooks into coding agents in five minutes — amos_gyamfi · 2026-07-28
- Ctrlb-decompose strips log noise before sending context to LLMs — ruhani_grover · 2026-07-28