Launched this week
AgentID
Shared memory, identity and tasks. Lower token costs.
27 followers
Shared memory, identity and tasks. Lower token costs.
27 followers
AgentID turns isolated AI tools into a coordinated team. Connect Claude, ChatGPT, Cursor, Codex, OpenClaw, and any other (we mean ANY:) Your agents keep memory across sessions, know what others already learned, collaborate through shared missions and handoffs, and reduce repeated context to cut token costs by up to 65%. Monitor runs, tool calls, memory updates, and savings from one live command center.
Hey PH! 👋
I built AgentID after getting tired of re-explaining my agents every time I switched tools, sessions, or environments.
The core idea:
Agents shouldn’t be stateless. They should have identity, memory, and continuity.
So I built a layer that gives them exactly that.
Here’s what AgentID unlocks:
🧠 Shared memory
All agents under the same identity read and write to a shared memory.
What one agent learns, all others instantly know.
🗺️ Memory map
A live graph showing how knowledge connects across agents and identities.
You can literally see how your system “thinks.”
📡 Studio (live monitoring)
Watch agents communicate, track their actions, and see token usage in real time.
No more black box.
🧬 Identity layer
Define personality, tone, and behavior once — and every agent stays consistent across all tools.
🔌 Works everywhere
MCP tools (Claude, Cursor, Codex…) and autonomous agents (OpenClaw, NanoBot…).
One command setup. Identity loads automatically.
—
Agents today forget everything, act inconsistently, and are hard to control.
AgentID makes them persistent, coordinated, and observable.
Curious what you think - and feel free to check my identity 😄
https://agentid.live/share/hexis
I appreciate product hunt users so much that I'm ready to give 99% discount for the Team plan, only for Product Hunt folks, enjoy :)
@artem_baygot Hi. One identity + shared memory for every agent = does it work across all tools?
@julia_zakharova2 Hey, great question - yes, that’s exactly the point.
The identity isn’t tied to a single tool. It works across any agent environment: Claude, Cursor, OpenClaw, Nanobot, you name it )
By using the same identity and shared memory layer, every memory written by one agent is instantly available to all. They share context, goals, and history - so they can truly work together toward the same outcome without you having to repeat yourself.