Dashboard

Open the dashboard and you land on one screen: what’s working, what needs attention, and what to inspect first. Every section adapts to how much data the account has sent — a brand-new workspace sees an activation path, not an empty chart.

What you see

A first-run account never sees this page at all: with no API key (and not in demo mode), the dashboard redirects straight to onboarding instead.

Once there’s something to show, Home leads with one conclusion, at most three priorities and one line on how much of the evidence is there. Everything else — follow-ups, the four area cards, the worklist and spotlights — sits in a collapsed Follow-ups and portfolio detail section below.

The header carries Share / PDF (see Executive summary) and Ask Morse, which starts with Ask about this page: a question about Home, ready to edit — see Ask Morse.

Org Brief

A single-claim headline sitting above everything else. It is hidden entirely rather than show a loading spinner or a broken state — no brief, no lifecycle line, or a failed request all leave the space empty.

Early in an account’s life it shows a canned line for the account’s current stage, with its call to action always a real link, never dead-end prose. Once the account has a steady verdict to report, it switches to one conclusion that Morse writes from your data: an overline This week’s brief · {window} with a status chip (for example Needs a decision or Healthy), the headline, and one action button. Open weekly summary → opens the Weekly Summary. Beside it, What changed lists the week’s changes in words (for example “spend +12% vs prior 7 days”), with All changes →.

When the top item is one agent failing runs, the headline names the agent, the share of its labeled runs that failed, and what those failures look like — a fixed category such as Provider billing (credits or quota), Rate limited by the provider or Calls timing out, taken from the agent’s costliest error group. It describes the failing calls; it does not claim why they failed. The raw error message never appears in the headline: open the error group to reveal it. When nothing needs attention, the headline is the top entry of your ranked actions, written as Do this first: followed by the action and its estimated dollars per month.

Priorities

Under the brief, Priorities lists at most three items, ranked. Each shows its title, a short detail, Owner: and a {next action} → link. The owner is the agent’s recorded owner, or No owner named — Morse never guesses one. When more items exist, the heading adds Top 3 of {n}.

The first item says Why this is first: — for example “It is rated high; the next item is rated medium.”, “Both are rated high; this one has newer evidence.”, or “A spend sign-off is waiting on a named person, so it stays first until they decide.” When nothing in the ranking separates two items, it says so.

Evidence line

One line says how much of the evidence Home stands on: “Evidence: 840 of 1,200 runs have a recorded outcome (70%) in the last 7 days. Runs without one are unknown, not failed.” See what evidence is missing → opens the Executive summary. With no runs, it reads “Evidence: no agent runs in this window yet.”

Follow-ups and portfolio detail

A collapsed section, closed by default, holds the rest of Home: Follow-ups, the Operating Summary cards, Needs attention and first cost findings, and Spotlights. Open it to see the sections below.

Operating Summary

Four cards, always titled Cost, Quality, Reliability, and Context & Memory, sorted so the most urgent (Action needed, then Needs attention) comes first.

Each card carries a capability-group chip, a status chip, a hero metric, a one-line reason, its own trend visual, optional sub-signals, and a stable call-to-action link. A card that doesn’t yet have enough samples shows a gate line explaining what unlocks it instead of a verdict.

The status chip reads Healthy, Worth watching (amber), Needs attention (red; earlier labeled “Action needed” and “Blocked”), Baseline forming or No recent data. Hover or focus the chip to see the measured fact behind it and the rule that produced it. Each status is defined in the glossary.

Needs attention

One merged, ranked, deduplicated worklist — every row is typed win, leak, risk, or next step, each with a severity badge and its own evidence link.

The panel also carries a persistent “N evidence items · view →” line at the bottom regardless of worklist state, and a header “Show all →” link into the failing-traces slice for the same window.

Spotlights

A fixed, curated band of up to five tiles — not a live-ranked list. Fleet context growth, this window’s prompt-cache savings (scoped to your top-spend agent and labeled with that agent’s name), auto-eval quality trend, and cost trend with deploy markers are always shown; a fifth, full-width Investigator tile appears only when a completed investigation exists in the window.

Follow-ups

Inside the collapsed section, the Follow-ups panel lists the findings your team is working on, overdue first, each with its next step, owner and revisit date. Open one to record or revisit a decision. Recording a decision runs nothing. See Follow-ups and decisions.

Weekly ranked performance brief

Read your weekly performance brief (1:11, narrated)

Open Weekly Summary (/home/weekly-brief; on Home, click Open weekly summary) for a rolling seven-day brief. It is a separate screen from this dashboard, generated when you open it. The Monday email is a different, shorter brief — the ranked worklist with evidence links — which you opt in or out of under Integrations → Notification Routing.

The brief’s header reads Weekly summary · {workspace} · {window}. Panels, in order: This week, a Trend line, Actions (ranked by estimated dollars), Decisions and results, failure groups, Last week’s changes, and Agents, most urgent first.

Trend covers up to five rolling 7-day weeks: “Over the last 5 weeks: Spend $410 → $388; success rate 81% → 84%.” Success counts labeled runs only, and a week shows a rate once it has 20 of them. With too few earlier weeks: “Not enough earlier weeks with data to show a trend yet.”
Decisions and results lists each decision a person confirmed in the window: its next step, owner and date, the change it led to (“Change: … on …” or No change linked yet), and the result, or Result: Still unknown. Results are the change tracker’s own readings, never a Morse estimate.
Last week’s changes names each change in words, not revision IDs, with Confirmed, Not confirmed or Rolled back {date}.
The success figure reads Success rate (runs with a reported outcome) — or “…reported or inferred outcome” when an outcome model labeled runs.

The brief lists active agents most urgent first, with links to their detail pages. Outcome verdicts compare each agent with its own labeled history. Spend is a separate column and never feeds the success verdict; sampled-quality evidence remains a separate lane elsewhere in the product. Unlabeled runs count toward activity and coverage, not success or failure.

Agents are ranked by verdict first, then by dollars at stake this week (spend on failing runs plus the estimated savings of that agent’s open actions), then by the larger drop in success rate. Dollars only order the list; they are never combined with the verdict into a score. The headline always names the agent to look at first and why (“data-extractor needs a look first: Calls timing out.”), and the brief ends with the top three ranked actions.

When there are open decisions to look at again, the brief adds Follow-ups due and What we still don’t know. See Follow-ups and decisions.

Pick a past week from the week menu to read the brief exactly as Morse wrote it that Monday — past weeks are frozen, not recalculated, so history starts from the first Monday after your agents send traces. Copy as Markdown puts the brief on your clipboard for Slack or a doc. Owners and admins can Share a read-only link: it has no trace links, expires after 30 days and can be revoked. The line under the title says whether the Monday email is on for you. Executive summary → opens the fixed-period summary.

When labeled outcomes or baseline history are insufficient, the success rate and verdict are withheld and the labeled-run coverage remains visible. An agent needs at least 20 labeled runs in the 7-day window to get a verdict; below that the row reads Not enough labeled outcomes. With 20 or more labeled runs but too little history, the row reads Baseline forming, and so do the Agents list and the agent page: Morse compares the latest day with the earlier days that have labeled runs and needs three of them, so labeled runs on at least four different days. With no traces in the window, the brief shows Nothing to report yet. Week-over-week changes are withheld when there is no prior evidence to compare.

Executive summary

Executive summary (/home/executive-summary) is a fixed-period report for a board or a management review, built from the same measurements as the rest of Morse. Pick a Period: one UTC calendar month — the last complete month by default, the month in progress (marked (so far)), or any of the six before it. A finished month is frozen the first time it is read and never recalculated; the month in progress is live until it ends.

The header reads Executive summary · {period} · {workspace}, then one verdict and Do this first: with its action — for example “2 follow-ups are due.” → “Open the decisions that are due and record what the result was.” Sections, in order:

Business workflows — each workflow’s purpose and owner, as your team wrote them (Name owner and purpose). Morse does not infer them.
Completed jobs and success — completed jobs (agent runs) and Success rate (runs with a reported outcome), each with its coverage.
AI spend — total spend, budget state and projection, and Cost per business unit, with an Other / unassigned row.
Reliability — material risks and open incidents your team recorded, including ones opened from an alert. Morse never opens an incident on its own.
Decisions and results — the same panel as the Weekly Summary, plus follow-ups due.
Changes since the last review, compared with the prior month.
Missing evidence and exclusions — outcome coverage, cost exclusions, workflows with no owner, a missing AI-spend budget, and unresolved comparisons, each with its next step.

The summary follows three rules, stated on the page: “Sampled quality checks and confirmed outcomes are different measurements and are never combined. Realized results are never added to candidate savings. There is no single health score.” Runs without an outcome are unknown, not failed.

When the prior month had too little activity, the comparison is described in words instead of as a percentage — for example “September 2026 had too little activity to compare ($3.10 across 12 runs).” A prior month needs at least 100 runs and $25 of spend (or 100 runs with an outcome, for the success rate) before Morse shows a change.

Share / PDF is on the Executive summary, in Home’s header and on Cost. It has its own period picker and two ways to share:

Download PDF — for members of this organization; it is not a public link. Any member can download it.
Create link (owners and admins) — a read-only link anyone with it can read. It has no trace links, expires in 30 days, and can be revoked. Other roles see “Only an owner or admin can create a public link.”

In the demo workspace the share link is turned off, and a demo PDF is marked as sample data.

How to use it

Act on Do this first and the Priorities; each carries its owner and next action.
Open Follow-ups and portfolio detail for the full worklist, cards and spotlights.
Use Share / PDF to send a month’s Executive summary to someone outside Morse.
Click the Org Brief’s action link to jump straight to the evidence behind its headline claim.
Click any Operating Summary card’s call-to-action to open that capability’s own screen.
Click a Needs Attention row’s link to open its evidence, or “Show all →” for the full failing slice.
Click a Spotlights tile’s title to open its home screen, or “All agents →” for the full roster.

Apart from naming a workflow’s owner and purpose on the Executive summary and creating a share link, this page is read-only — every other interaction is navigation.

Before you have data

Each section is honest about insufficient data rather than filling space with a number it can’t stand behind:

No runs in the window: the evidence line reads “Evidence: no agent runs in this window yet.”
Not enough runs yet: the Org Brief and any under-threshold Operating Summary card show a gate line naming what unlocks them, instead of a verdict.
Nothing needs attention: “Nothing needs attention right now” / “No evidence-backed wins, leaks, risks, or next steps for this window.”
A Spotlights tile without enough data shows its own honest placeholder, e.g. “Unlocks once an agent has run in this window.”

To get your first data on this screen, follow the quickstart.

Traces — the runs behind every Needs Attention and Spotlights row.
Agents — per-agent detail behind the Operating Summary and Spotlights cards.
Cost — the full picture behind the Cost card and the Cost trend tile.
Reliability — the full picture behind the Reliability card.
Client reports — scheduled reports for your customers and your team.