Launching today
Nysa
Shared memory for teams that work with AI agents
1 follower
Shared memory for teams that work with AI agents
1 follower
Nysa is shared memory for teams that work with AI agents. Meetings and email become profiles, knowledge and skills, each fact with its source. Every teammate's agent reads and writes it over MCP, and sees only what its person can see: scopes, private projects and scoped API keys.
Hi Product Hunt, I'm Javier, the maker of Nysa.
I started Nysa after building a lot with AI agents and harnesses like OpenClaw. One agent for one person is great. But the moment a team uses agents, everyone's agent knows a different slice of the company, and nobody's agent knows what was decided in the call they missed.
Sharing everything isn't the answer either: not everyone should see everything.
Nysa fixes both. A meeting notetaker, Gmail and Google Calendar feed one company brain of facts, each with its quote, date and source. Every teammate's agent (Claude, ChatGPT, Codex, Cursor, OpenClaw, Hermes) connects over one MCP server, signed in as that person, and sees only what they can see. Raw email stays private to its owner; the team gets only the facts. Nysa keeps a living profile for every person and company, a shared knowledge base, projects, and skills: a workflow one agent learns, every teammate's agent can reuse. Agents write back too, and API keys can be limited to specific scopes and projects.
Setup takes under five minutes. Plans are per team, not per seat: $99/month up to 10 people, with a 14-day free trial.
It works with Google accounts only for now. I'd love to hear what you'd want your agents to remember that they currently forget, and where the permissions model doesn't match how your team works. I'll answer every comment today.