One memory for every AI your team uses.
Shane captures the decisions, context and skills behind your work, keeps them organised and current, and hands them to every agent and teammate who needs them.
Shane is in development. This page describes what we’re building; join the waitlist for early access.
Context saves itself.
Legal want another pass on the terms before we ship.
Agreed. Moving launch to 21 Oct so legal has the week.
Saved to Shane as a decision
Slack, Notion, Google Drive, Linear, GitHub and meeting notes feed Shane. No new place to write things down.
When something is agreed in a thread, a meeting or a doc, Shane saves it as a decision.
Every entry keeps the reason, the owner, the date and a link to where it came from.
Mark a message, a doc or an AI answer as worth keeping. Shane files it.
Meeting notes are turned into the decisions, owners and follow-ups that came out of them.
When a decision changes, Shane links the new one to the old and stops serving the outdated version.
One brain, always current.
One place with four views: what’s true, what was decided, how work gets done, and who knows what.
A running history of decisions by team and project, so you can see how you got here.
Context is grouped the way your company works, so each agent and teammate gets what’s relevant.
Find the decision, the reason or the person behind anything your team has worked on.
Every agent starts with your context.
- Annual plans are 20% off Maya · Pricing review
- Monthly price stays the same
- Launch is 21 Oct Tom · #launch
Adding the annual plan at 20% off, per Maya’s 3 Oct decision. Monthly is unchanged.
+ { id: "annual", interval: "year", discount: 0.2 }, { id: "monthly", interval: "month" },
ChatGPT, Claude, Cursor, Gemini and your own agents read from the same memory.
Shane speaks the Model Context Protocol, so any MCP-capable assistant or agent can use it.
Agents get the relevant decisions and rules up front, instead of you pasting them in.
Each piece of context links back to the message, meeting or doc it came from.
Ask what was decided, why, and by whom. Get an answer with the receipts.
Read and write context from internal tools and custom agents.
Write it once. The whole team has it.
- Pull merged work since the last release
- Group by customer impact, not by team
- Plain words, no internal codenames
For customers Annual plans now available at 20% off.
Fixes Checkout no longer double-charges on retry.
Turn the prompt or process that works into a skill with steps, inputs and examples.
Skills show up where your team works, like a slash command in a teammate’s assistant.
Every skill has an owner, a version history and a reminder to review it when something changes.
Know which skills the team relies on and which have gone stale.
See how decisions move through the company.
What was decided across teams this week, by whom and why, in one read.
Shane flags when two teams start solving the same problem.
See which context and skills your agents are working from, without reading anyone’s chats.
Your context, your rules.
Shane follows the access rules of the tools it connects to, for people and for agents.
Pick the tools, channels and teams it connects to. Exclude anything sensitive.
Remove a decision, a source or a whole connection, and it’s gone from every AI.
See which context went to which agent or teammate, and when.