My Decisions, a Runs ledger with receipts, and a Live Feed of your agents' work
Everything your agents are waiting on now sits in one ranked queue, every agent run lands on a calendar with a receipt, and a Live Feed tells you in plain sentences what your agents did. A command palette, a Task board with true totals, approval and send limits for agent email, connection health and a What's new panel round out the release.
This release adds the screens you need once several agents work for you at the same time: what is waiting on your decision, what ran and what it cost, and what happened while you were away.
My Decisions reads your Inbox as a ranked queue. Runs puts every agent run on a Day, Week or Month calendar with a receipt for each one. Live Feed turns the activity history into plain sentences you can filter by agent and kind.
Around them, a command palette reaches any screen or record from the keyboard, the Task board counts every Task and puts urgent work first, agent email can wait for your approval within send limits, MCP connections show their health, and a What's new panel in the top bar lists product updates. Most of these are first versions, and each entry says what is not there yet.
On this page
Features(6)
- New featureInbox & decisions
My Decisions: one ranked queue for everything waiting on you
Every question, approval and escalation your agents are waiting on now sits in one list, with the ones holding work back at the top. Answer one and you are told what happened to the work.
Open Inbox and choose the My Decisions tab (
/inbox?view=decisions). It lists questions, approvals and escalations, never notices.- Ranked for you. In Open, decisions that hold work back (the run is paused or the Task is blocked) come first, then by how sure the agent was, then oldest first.
- Filter and find. Switch between Open, Answered and Archived, filter by kind, search, or narrow to one Agent, Task or Mission. Filters stay in the link.
- Context before you answer. Each decision shows why it is blocking, what the agent tried, any risks, and links to its Task, Mission and Agent. j and k move between decisions.
- See the outcome. After you answer, a line says whether the agent picked the work back up, your answer went to a run that is already going, the work is queued, or it could not restart, with a link to run the Task yourself.
In this view, rejecting an approval or choosing something other than the recommended option asks you why. A decision answered on Home or on a Task page is closed here too, and Home's approvals block links in with See all decisions.
Limits: the queue loads 25 decisions at a time behind Load more, and an answer cannot be undone yet.
- New featureActivity & runs
Runs: every agent run on a calendar, with a receipt for each
A new Runs page lists every agent run by day, week or month with totals for that window, and each run opens a receipt showing what it did, what went wrong and what it cost.
Open Runs in the sidebar, just above Activity (
/runs).- Pick a window. Switch between Day, Week and Month and step back or forward. Times follow your profile's timezone, and you can go back 12 months.
- Filter. Narrow by Agent, trigger and outcome, or search run summaries and errors.
- Totals for the window. A side rail shows runs, success rate, errors, agent time, spend and tokens for exactly what is listed. Click an outcome figure to show only those runs.
- A receipt per run. Open a run to see its summary, the exact error with secrets removed, its cost (settled total, metered usage per capability and model, and credits), message and tool call counts, files touched, the Knowledge Base documents it cited, and the Mission, Task, Work, Agent or schedule it belongs to.
The receipt also appears on each session's detail page, and a run's cost under Top runs on the Costs page opens its receipt. A figure the platform did not measure shows as a dash or Not reported, never as zero.
Limits: the split between input, output and cached tokens is not reported yet, and filtering by Mission, Work or model is not on the page yet.
- New featureActivity & runs
Live Feed: what your agents did, in plain sentences
The Activity page has a new Live Feed view that reads your activity history one line per event, naming who did it, with filters by agent and kind and a link to what each line is about.
Open Activity and switch the view to Live Feed (
/activity?view=feed), next to Log and Schedules.- One sentence per event. Each line names who acted (an agent, you, an outside source or Ever Works), what happened, a kind label and how long ago. When there is something to open, such as a run, Task, Mission or Work, the line links to it.
- Filter. Pick agents from chips (up to 20 at once), choose kinds (work, decision, delivery, problem, system), or turn on Only failed. Filters stay in the link and come back on your next visit.
- Scroll back. Older entries load as you scroll, or with Load older.
- Keyboard. j and k move between lines and Enter opens one.
The feed reads the same records as the Log, so nothing is recorded twice. A new line keeps the agent's name from the moment it happened, so renaming an agent later does not rewrite its history.
Limits: the feed checks for new entries every 10 seconds instead of streaming them, and one visit reaches back at most 20 pages. Entries from before this release show the agent's current name, and anything older than 90 days stays in the Log.
- New featurePlatform
Jump anywhere with the command palette (Ctrl+K or Cmd+K)
Press Ctrl+K or Cmd+K anywhere in the dashboard to search your workspace, jump to any screen or run an everyday command without leaving the keyboard.
Press Ctrl+K (Windows and Linux) or Cmd+K (Mac), or click Search… in the top bar. / also opens it when you are not typing in a field.
- Search your workspace. Type two or more characters to find Missions, Tasks, Agents, Works, Ideas, Skills, Teams and Knowledge Base documents in the active workspace, grouped by type.
- Go to any screen, including every Settings page and the pages of the Work you have open.
- Run a command: create a Mission, Task, Agent, Goal or Meeting, switch workspace or Work, open the AI chat, copy a link to the page, toggle dark mode or sign out.
- Move fast. Tab narrows the list to one group, Ctrl/Cmd+Enter opens a result in a new tab, and your recent items show while the box is empty.
If nothing matches, the palette offers to ask the AI chat or to create a Task from what you typed. Inside a Work's Knowledge Base, Ctrl/Cmd+K still opens that workbench's own search, and the top bar button opens this palette.
Limits: runs, decisions, Memory, Goals and Meetings are not searchable yet, and recent items are remembered per browser, not across devices.
- New featureAgents
Agent email can wait for your approval, within send limits you set
Choose per Agent whether its email goes out on its own or waits as a draft for you to approve, and cap how much it may send. Both are enforced on the server for every message.
Open an Agent and choose the new Inbox tab. The Sending policy panel has two parts.
- Approval. Draft for review keeps everything the Agent writes as a draft until a person approves it. Auto-send lets it send on its own. Waiting drafts show in the Agent's Inbox with Approve & send and Discard, and in your approvals queue. A draft is sent once, and approving in bulk never sends email.
- Send limits. Set sends per 24 hours, sends per minute, recipients per 5 minutes and recipients per message. Leave a field blank to inherit it, or enter 0 for no limit. Apply recommended limits fills them in one step, and a live meter shows usage against each limit.
Limits are checked at the moment of sending, whether the Agent, another Agent or a person wrote the message. A message over a limit is not sent: a draft stays a draft, and you are told which limit was reached and when capacity returns. The same tab lists the Agent's messages and lets you choose which of your addresses it sends from and receives at.
Limits: an Agent you have not configured keeps sending as before, with no limits. Adding a new email address is not in the product yet; that step still goes through the API.
- New featurePlatform
A What's new panel inside the app
A What's new control in the dashboard top bar shows how many product updates you have not read yet and opens a panel that lists them, with filters by area.
Look for What's new in the top bar, next to the notification bell. A badge shows how many updates you have not read.
- Read in a side panel. Updates are listed newest first, with any pinned update at the top, each marked New, Improved, Fixed or Security. Some include a button that takes you to the screen they describe.
- Filter by area: Agents & Missions, Decisions & Safety, Knowledge & Memory, Connections & Computers, Runs & Costs, or Platform.
- Your own read state. An update counts as read once it has been on screen for about a second, or use Mark all as read. It is tracked per person, so switching workspace does not change it, and a new account starts with nothing unread.
The updates you see are the ones that shipped with the version of Ever Works you are running, and the panel never opens by itself.
Limits: the panel holds a short list of selected updates, not the full changelog, and the badge count refreshes on your next page load.
Improvements(2)
- ImprovementTasks & goals
The Task board shows true totals and puts urgent work first
The Kanban view of Tasks now counts every Task in each column, orders cards by priority, and keeps its view, sort and date window in a link you can share.
Open Tasks and choose Kanban (
/tasks?view=board).- True column totals. Each header counts every Task in that status under your filters, not only the cards loaded on the page. Show N more loads the next cards of that one column.
- Urgent work first. Stalled Tasks in progress come first, then priority P0 to P4, then the oldest update. Use Sort to switch back to Recently updated.
- Choose how far back completed work goes. Completed shows Done and Cancelled Tasks from the last 7, 30 or 90 days, or all time.
- Linkable. The view, sort and completed window stay in the link, the page remembers the view you used last, and applying a filter keeps you on the board.
If the platform refuses a move, the card goes back to its column and the reason is shown. Dragging a cancelled Task explains that it cannot be reopened.
Limits: Cards is still the view you land on by default, and a stalled Task is not flagged on its card yet.
- ImprovementConnections
See which connections need attention, and give agents read-only access
Settings → Connections now shows the health of each MCP server with a banner when something needs attention, and an Agent's Capabilities tab can limit a provider such as GitHub to read-only access.
Connection health. Open Settings → Connections. Each MCP server now shows Healthy, Having trouble, Expired, Unreachable or Not checked yet, based on its last real connection: a run using its tools, a tool call or Test. Credential problems come with a one-line fix, and a banner counts the connections that need attention.
Access levels. Open an Agent, go to Capabilities → Access levels, and choose Read only or Read and write for a provider. Read only lets the Agent look things up but never change anything. For GitHub, Read and write adds committing to a repository and opening pull requests. The change applies on the Agent's next tool call. If a broader setting narrows the level, the section says so, and a level the connected account does not permit asks you to reconnect.
Credentials only over https. Saving a connection that carries an auth header now needs an https address, so the header is never sent in the clear. One saved earlier on plain http is marked Unencrypted and is refused the next time it tries to connect.
Limits: GitHub is the only provider with access levels so far. Health updates only when something connects, with no background check yet.
Plus 4 fixes in the GitHub release notes (opens in a new tab).