Console menu guide

Saving and searching happen in your AI; the console is where you build and maintain the ways in. What each of the six menu groups does.

Saving and searching are done by your connected AI, with ak. The console (app.aiakiv.com) is everything around that. creating teams and projects, changing the save target (Main), and inspecting and managing what's accumulated. This page maps what each console menu does.

The one-sentence principle. The console is for setup, management, and checking. Do the day-to-day saving and searching in your AI, and think of the console as somewhere you drop into occasionally to maintain the ways in.

Much of what follows can also be asked of your AI: creating a team, inviting someone, link invite codes, reading the audit trail, managing projects, switching the save target, managing cards. Reads work from the start; anything that changes something works only where you turned it on in Reference → AI permissions. Deleting people, teams and accounts stayed on screen only → Asking your AI to do console work. Asking the AI works only on a connection that has app features. On a connection without them, such as the official ChatGPT plugin or a Claude account connector added with an address that lacks ?apps=all, you do it in the menus below yourself → The app action tool.

Layout

Top bar: the AiAkiv logo on the left (goes to www.aiakiv.com), and on the right team picker, language, and the account menu (email ▾).
Left sidebar: the six menu groups below. On mobile, open it with the ☰ at the top left.

The menu groups are Teams · Projects · Data · Operations · Support · Reference.

Teams: managing the owner of memory

Memory is owned by teams (see Teams and projects). This group handles them.

Teams: the teams you belong to, including the personal team created at signup. Explore (public read) teams can also be attached here. You connect read-only with no invitation or membership, and to write you use a project in your own team.
Team structure: shows who can read what inside the team as a tree (the permission tree). Admins assign and remove members up and down it here, and where needed move a memory's access scope (its ACL node). Author and tags don't change.
Audit log: recent permission governance actions. Admins see the whole team; other members see only their own. Come here to retrace "who changed permissions, and when."
Links: make and cut the ways in to another team's memory. Visible only to the team owner. Proposing and accepting, checking the 'readable now' conditions, and tuning aliases (name connections) all happen here. The public attach that ties a public memory into your team is on this screen too → Links · Public memory.
Link history: the record of when a link was made, changed, and cut. It's kept for both teams: it makes no sense for your access to be cut without the affected party seeing it.

Projects: the ways into team memory

A project is a way in to team memory (not a warehouse).

New project: create a way in. Set the name, domain, group (sub-folder), persona (the role the AI takes on this way in), and TTL, and get the MCP config snippet and API key to attach to your app here.
Project management: the list of projects you've made. Set a project as Main to change where saving currently goes (the active target), copy MCP config, or delete a project. Switching the save target is done here by a person by default. An AI can switch it only if you turned on the save target switch permission in Reference → AI permissions. Each project card also issues an API key pinned to that project.
- Turning Main off and on: "Turn Main off" in this screen's save target line turns Main off for a while. While it's off, connections that follow Main (account connectors, API keys not pinned to a project) refuse every request, saving and searching included, and only folder connections and project-pinned API keys work. While it's off, the save target line shows the name of the Main you turned off with "Turn Main on", and that project's card button in the list also changes to "Turn Main on". Pressing it brings back the Main you had just before, unchanged. Pressing "Switch to main" on another project moves Main to that project and turns it on. Turning Main off and on is done here by a person by default; with "Turn Main off" and "Turn Main back on" on in Reference → AI permissions, you can ask the AI too → When the account connector and a folder connection are both attached.
API key management: issue and re-issue per-project API keys. A key is shown in full only once, at issue time, so copy it then. Re-issuing invalidates the previous key. To let an AI connected with the key edit through the console, check "Allow console management by AI (manage)" when you issue it (project management's key issue has the same box) → Authentication.

Data: inspecting what's accumulated (read-only)

Everything here shows only what you can read. What you see in the console is for checking; the actual searching is easier from your AI (Searching).

Summary: the node tree you can read in the current team and the number of memories (events) per node.
Search: search by meaning within your read scope. Each result shows the node it's saved in.
Browse: page through memories in your read scope in save order.
Analyze: see frequently occurring entities and their cohesion. It's meaningful once some memory has accumulated (Core concepts).
Edit: edit and merge entities and tags directly in the console (four tabs: entity edit, entity merge, tag edit, tag merge). It's the curation tool for finding duplicate entities and merging or retiring them. Team owner and admins only.
Cards: manage the public cards you had an AI create (list, copy address, images, delete, search visibility). Creating happens in your AI, not here. Cards here are already public → Cards.

Operations: usage and moving data

Usage: events used against the limit, per team. Teams you own appear here.
Billing: your current plan and upgrades. Checkout and invoices open in the billing portal. Enterprise is by arrangement (contact us).
Export / Import: move the knowledge graph in and out of AiAkiv. Export one of a team's domains in the open-source MemoryWeft format (.mweft.tar.gz), or import self-hosted MemoryWeft data wholesale (events, entities, tags, edges, embeddings). Team owner and admins only.

Support: contact

Contact: leave a question and you get a reply by email. Write in whatever you're stuck on, or a bug.

Reference: announcements, install, usage, settings, AI permissions

Announcements: service notices and urgent bulletins. New notices also run as a banner at the top of the screen, and urgent ones pop up once. Past notices are collected in this menu.
Install: how to connect AiAkiv to the AI you use (Claude, ChatGPT, Cursor, Gemini, and so on). Gives per-app config and copyable snippets. (Same content as the setup guide on the web.)
Usage: basic usage guidance for saving, tagging, searching, and chained saves. You can also have your AI walk you through it step by step.
Help: a searchable reference of AiAkiv's terms and concepts (memory, team, project, save target, event, entity, tag, persona, MCP, and more). Type a word in the search box at the top and it filters live.
Tips: the tips page collecting the click-through learning slides (opens in a new tab).
Demo: the demo tour where you ask questions of a sample built from a real development record (opens in a new tab).
Settings: change the theme (dark/light) and the language. Receive tips in AI chats is also turned on and off here. When it is on, AiAkiv now and then adds a one-line how-to note to an AI tool response (at most twice a day).
AI permissions: turn on and off, permission by permission, what your connected AI may do in your account through the console app. Reads start on, edits start off, and a permission with a risk mark asks you to confirm when you turn it on. "Turn Main off" and "Turn Main back on" in the Save target group let the AI turn Main off and on. They are turned on separately, so you can leave turning it off to the AI and turn it back on yourself → Asking your AI to do console work.
Danger zone: deleting a team and closing your account (full account deletion). Everything on this page is permanent and cannot be undone.

AiAkiv logo (left): goes to www.aiakiv.com.
Team picker: choose the team you're working on (the Data and Projects screens are scoped to it).
Account menu (email ▾): theme and language, API key management, Danger zone, sign out.
AI permission notices: with the AI's save target switch permission on, the top bar shows "AI can change save target"; when the permission settings cannot be read, it shows "AI permissions unavailable". Click either to open AI permissions. While Main is off, the "AI can change save target" notice is hidden, because the AI's switches are refused during that time even with the permission on.
Main off notice: with Main turned off, the top bar shows "Main off". Click it to go to Project management.
Mobile (screen width 760px or less): "Main off" appears not in the top bar but as a one-line banner just below it: "Main off · Main connections paused. · Turn it on ›". Tapping the banner doesn't turn Main on by itself. It takes you to Project management, where you press "Turn Main on".

Common confusions

"Saving doesn't change." → The save target (Main) is changed by a person in Project management. An AI can change it only if you turned on the save target switch permission in AI permissions; the save and search tools cannot change it (Teams and projects).
"I can't see a memory." → Check the team you currently have selected and your read scope. The Data screens show only what you can read (Common problems).
"Can't I save from the console?" → Saving and searching are done in your connected AI, with ak. The console is where you create and manage the ways in.