Launched this week
Context Engine
One memory for your AI agents and human team members.
101 followers
One memory for your AI agents and human team members.
101 followers
Context Engine is one memory for your AI agents and your team. Claude, ChatGPT, Cursor and Claude Code save to it as they work, and the people on your team read it, ask it questions and add files. Free to start.
Free Options
Launch Team
ArticosTest your messaging on your audience, right inside Claude or ChatGPT
Promoted
This is useful, when we want to track someone's progress in the team, if they are doing some things wrong. Giving contexts to multiple agents does not make sense for me as I have already planned in detailed information from start of the development. If I want to use another agent for review, I should not give it the same context as it will just agree to the decisions it have taken with another AI. We do not want that in this condition. I cannot think of a scenario where I would need different AI to do work but I think there will be if we would have AWS AI or any integration AI which can just automate the changes we did on our development.
With multiple users adding to it, are there any kind of guardrails for either intentional or unintentional changes that are wrong, or full of errors? Or a way to set up a hierarchy so that certain users are considered primary and their contributions take priority over others?
One idea for that could be a 2nd-user check - new context that would affect other people gets routed for review first. Cool idea, interested to see how it works!
Being able to read the same memory your agents use seems useful. I'd want to see what they saved before the next agent picks up the task.
Congrats on Context Engine. One memory that the agent and the teammate on the other end both read is the part that stuck.
Congrats on the launch! Sharing one memory between agents and humans is the part I find hardest to get right. When an agent writes something stale or wrong, how do you keep it from quietly becoming "truth" for the whole team? Do you have provenance or decay built in, or is it on humans to curate?
our agents' rules live in a file someone edits by hand, and the hard part is cleaning out what's no longer true. when a teammate corrects something an agent saved here, does that agent stop saving it again?