Engramic

Your organisation keeps forgetting what it already worked out.

Decisions get made in calls, chats and threads, then lost. A month later someone makes the same call again, or the opposite one. A new colleague, a client or an AI agent starts from an old document, and nobody can say what was actually agreed.

Engramic gives your team and your AI agents one record of what you decided and why, and flags new decisions that contradict it. It's built on three ideas: Memory, Context and Consistency.

START HERE
See what needs doing, and what changed. Actions coming due, what your AI recorded in your name, open questions and a catch-up on everything since you last looked, on one page.

MEMORY: WHAT HAVE WE DECIDED?
Record decisions, goals and constraints as the work happens, with the options you weighed and why you chose one. Decisions can say who agreed them. When a decision is replaced, the old one stays on the record with the date it changed, so everyone can see which version is current.

YOUR AI DRAFTS THE RECORD. YOU PUBLISH IT.
Connect Claude, Claude Code or another MCP client in a few minutes. As you work, it drafts decisions and actions from the conversation you're already having. You check the wording and publish. Agents never count as seats, so connect as many as you like.

CONTEXT: WHY DID WE DECIDE IT?
Share a page as a link. Colleagues, clients and agents read it with its reasoning and history, and readers can ask the page questions and get answers that point to the section they came from.

Ask Alia, the workspace librarian, what was decided and why. She answers in plain English from your team's own records, with the reasoning attached.

CONSISTENCY: DOES THIS CONTRADICT WHAT WE DECIDED?
This is the part we think matters most. Engramic connects decisions across time and flags a new one that can't hold alongside an existing one: a launch date set before the audit it was meant to wait for, say. Open conflicts show where you already look, on Activity, the topic tree and the topic itself, and stay open until a later decision settles them. Engramic flags; people decide. Alia also raises gaps, like a goal with no constraint, and asks the right person.

YOURS, NOT YOUR MODEL PROVIDER'S
Nothing you record is used to train any model, ours included. Workspace data is stored in London. Because the record lives in Engramic rather than inside a model, you can change models without losing what your team and agents know.

PRICING
Free: one person, free forever. MCP tools with full history, unlimited agents recording on your behalf, agent-ready share links and Alia-lite.
Standard, £9.99 per seat per month: the full Alia agent, web search, external MCP connections, artefacts and templates.
Pro, £24.99 per seat per month: resolving conflicts with Alia, document analysis, your own model providers and priority support.
A seat is a person. Prices exclude VAT.

WHO IT'S FOR
Small teams and agencies already using AI for real work, who have watched a colleague or an agent act on a decision that had already changed.