13 · Your AI agents

The platform runs a workforce on your behalf: analysts that write documents, watchdogs that monitor your portfolio and the market, autopilots that act under a mandate you set, and the workflows your own team composes. This chapter explains the four classes, lists every agent, and walks you through building one. The one rule behind all of them: an agent must show its work. Every run is logged, every figure is grounded in your stored data, and nothing leaves the platform without a gate you chose.

The Agent Center

Where: Team & AI › Agents · /agents
Who: everyone

The operations desk over every agent working for your organization: who is on duty, what they did while you were away, and where every piece of work lands. The Center is an application with five views on one masthead, and it presents agents as a staff: a roster you manage, a bench you hire from, and a work ledger that proves the output. Press 1 to 5 to switch views and / to jump to Ask Vera; each view is addressed (/agents?view=roster), so a forwarded link lands on the right one.

The masthead

• Your workforce — the headline counts the agents on duty for you, and the line under it says what happened while you were away. Four figures: On duty, split between the standing agents and the ones your team built; Runs, 30 days; Waiting on you, which is runs paused until somebody decides; and Hours given back. Three controls sit beside them: Which agents do I need? (a four question interview that recommends a starting roster), Meet your staff (the introduction, replayable any time), and New agent.
• What "on duty" counts — the standing watchdogs and autopilots sweep your estate on the platform's clock whether or not you have rostered them. What the roster decides is whether they may tell you anything: an agent you have not put on your roster is silent on every channel, in-app, email, and chat alike. Nothing alerts you until you ask for it.

The five views, on the masthead's band

• Overview — the floor at a glance. Ask Vera leads: describe your situation in a sentence and Vera answers with a numbered shortlist of the agents that cover it (see Ask Vera which agents you need). Under it, the four departments, one tile per class with its verb: Analysts write, Watchdogs monitor, Autopilots act, Workflows compose. Then the latest entries of the work ledger and Today on the floor, the agents scheduled to run on today's date.
• Your roster — the agents you chose to lead with, plus every workflow your team built. Workflow cards carry a 14 day heartbeat sparkline and their run count. The tab's badge counts them.
• Hire agents — Start from your situation first: six named situations ("A renewal lands inside 90 days", "I want every invoice checked") that put the right bundle of agents on your roster in one click. Then the bench: every other agent, grouped by class, each one add away. A benched watchdog still sweeps your estate, but it reports nothing to you until you roster it, so the bench is silence rather than absence.
• The week — a week with your agents, read from the schedule the platform actually deploys, not from a label, so a day with nothing scheduled says so. Agents that run monthly, on an event, or only when you ask are counted in the line beneath rather than pinned to a weekday they do not keep.
• The ledger — everything your agents did, newest first: watchdog findings and workflow runs, each row wearing its agent's monogram, with a link into the full run history.

The rail, in reach from every view

• Waiting on you — appears only when workflow runs are paused on an approval, naming each run and how long it has waited, with a link to review and approve.
• What is being watched for you — the coverage card: price lists, billing, standings, anomalies, vendor health, and paper hygiene, each with its cadence and a link to the detail.
• How staffing works — the three rules in short: the sweeps run regardless, the roster is your alert list, and the autonomy dial sets how far an agent may go alone.

Every card shows the agent's monogram, class, and cadence, an arrow sentence (what it reads, what it does, where the work lands), an example find, a run receipt, and links to its dossier and its output page. Agents that can act (the autopilots) carry the autonomy dial: Watch, Draft, or Act, one vocabulary everywhere.

The run receipt is the card's proof of work: when the agent actually last ran, read from the platform's own scheduler heartbeats, and when it runs next, computed from its real schedule ("Last ran 2 hrs ago · next today at 06:00"). Event driven agents state their trigger instead ("Runs when you forward a vendor email"), and the on demand analysts say they run when you ask. When a last run cannot be derived yet, the card shows only the next run; a timestamp is never invented, and an agent nothing actually runs is not advertised in the Center.

How to use it

1. Scan the masthead; anything waiting on an approval is your queue, not the machine's, and it stays pinned in the rail on every view.
2. Read the ledger, on the Overview or in full on its own view, to see what ran while you were away, and open any row.
3. Not sure where to start? Tell Ask Vera what is going on, use Which agents do I need?, or pick a situation card on the Hire agents view; all three put a sensible bundle on your roster.
4. Use New agent to add to the workforce.

Related: Workflow agents · Vera AI

Ask Vera which agents you need

Where: Team & AI › Agents, the box under the workforce band · /agents
Who: everyone; putting agents to work needs edit access

Most people arrive holding a problem rather than shopping for agents. Say the problem in your own words and Vera staffs it: "our Microsoft renewal lands in 60 days", "I think we are being billed for seats we dropped", "leadership asks about software spend every month".

On the screen

• The question box — one line, with four starter situations underneath it until you have asked something.
• Vera's answer — two to four sentences on why these agents fit your situation and what will land in your inbox.
• The numbered shortlist — up to five agents, best fit first, each row showing its number, monogram, what it does, and how often it runs. An agent already on your roster is marked On roster and cannot be added twice. Every other row carries a checkbox, ticked by default.
• Put N on my roster — activates the ticked rows. The count follows what you tick.
• Built for you — appears only when nothing on the shelf covers what you asked for: Vera drafts a custom workflow agent, saves it, and shows you its chain of steps. It arrives switched off, so nothing runs until you press Turn it on. Edit every step opens it on the workflow desk.

How to use it

1. Type your situation and press Ask Vera, or tap one of the four starters.
2. Read the shortlist. Open any agent's dossier first if you want the detail before deciding.
3. Answer with the numbers you want, for example 1, 2, 3 and 4, or all of them. Ticking the checkboxes and pressing Put N on my roster does exactly the same thing.
4. The agents join your roster immediately and start on their next scheduled run. Tune the leash or bench any of them from the roster below.

Good to know

• Vera recommends only agents that exist and are genuinely running; she cannot invent one or offer one the Center hides.
• She never recommends an agent you already have. If you pick one anyway, she says it is already covered rather than adding it twice.
• Answering with numbers is resolved against the list on your screen, so the agents you name are always the ones you get.
• Ask Vera needs the AI plan. Without it, the four question front desk interview still recommends a starting roster.
• Viewers see the same recommendations but cannot switch agents on; ask an owner or a member with edit access.

Agent dossiers

Where: Team & AI › Agents › any agent · /agents/{agent}
Who: everyone

Every agent has a home page: what it reads, when it runs, what it delivers, and your settings for it, in the same record page grammar as the rest of the app.

On the screen

• The masthead — monogram, class, cadence, the agent's one line, and the See its work button.
• Highlights — what it reads (as chips), what it does, where its work lands, and your setting: roster membership and, for agents that can act, the autonomy dial.
• Example output — a curated miniature of the agent's deliverable, always labeled as an example, never your data.
• Recent findings — for watchdogs, what the agent flagged for you lately, straight from your notifications.
• Part of these situations — the situation bundles this agent serves.

How to use it

1. Open a dossier from any card, ledger row, or week strip pill.
2. Read the example output to understand the deliverable in ten seconds.
3. Add the agent to your roster, or set its autonomy, from the highlights panel.

Build an agent

Where: Team & AI › Agents › New agent · /agents/new
Who: everyone · building workflows needs edit access · drafting needs an AI plan

One wizard for all three buildable classes. Pick the job, and the wizard routes you: document writing goes to the custom analyst builder, monitoring shows you the watchdogs already on duty, and routine automation drafts a workflow from one sentence.

On the screen

• Three doors — Write documents for me (Analyst), Watch something and tell me (Watchdog), Do the steps I always do (Workflow).
• Describe it in a sentence — for workflows: state the goal in plain language and the AI drafts the chain: trigger, steps, and the approval gate where anything would leave the platform. Three template chips above the box (review every proposal upload, the weekly renewal memo, routing vendor emails) prefill a proven goal you can edit, so you never start from a blank box.
• Review the draft — the drafted chain as a step map and a numbered list. Rename it, change the trigger, or rewrite the goal and draft again. Full per-step settings live on the workflow desk after saving.
• Test with my data — runs the chain end to end without sending anything outside the platform, so you read the output before the agent earns a trigger.
• Turn it on — enables the agent and shows you exactly where its work will land.

How to use it

1. Pick the door that matches the job.
2. For a workflow: describe the goal in one sentence. A goal that names the vendor, the cadence, and the audience drafts best.

1. Press Draft the agent and read the drafted chain: the name, the trigger, the step map, and where the approval gate sits. Rename it, change the trigger, or rewrite the goal and draft again.

1. Save the agent, test with your data, read the run on the workflow desk, then turn it on.

Tips

• Drafts are conservative on purpose: anything that emails a person outside a pure internal note gets a Wait for approval step. Remove it on the workflow desk only when you trust the chain.
• A goal that names the vendor, the cadence, and the audience drafts best: "Every Monday, review our Microsoft position and email the owners a short memo when anything moved."

The four classes

Analysts write documents. On demand (the six playbook specialists plus up to six custom analysts your org defines) or on a schedule (the Monday vendor wire, the meeting pack before each vendor call, the monthly board report, the spoken monthly briefing). Their output lands in your documents library or on the page named on their card.

Watchdogs monitor and flag. They run on the platform's clock, read only the data named on their card, and surface findings on their home pages, in My work, and by the email digests you turn on. You do not build watchdogs; you tune what each one tells you from its own page.

Autopilots act under a mandate. The Renewal Autopilot works your renewal calendar at the autonomy you choose: watch and remind, draft for your approval, or negotiate under a mandate with fees only on invoice-verified savings. The negotiation ghost writer drafts vendor replies from forwarded email threads. Every action an autopilot takes is logged on the record it touched.

Workflows are yours. A trigger plus up to eight steps: AI analysis, classification, extraction, an if then gate, a wait for approval, then email, Slack, a reply to the sender, or a task. Build them in the wizard or by hand on the workflow desk.

The roster

Every standing agent the platform runs for you, the clock it runs on, and where its work lands. Workflows are not listed because they are yours: the Center shows each one your team builds, with its own trigger and run history.

Watchdogs

Autopilots

Analysts

Trust and guardrails

• Grounded output. Analyst and workflow memos read your live portfolio data; figures are checked against stored records, and missing facts come back empty, never invented.
• Approval gates. A Wait for approval step pauses any workflow run until a named human decides; rejections end the run. Autopilot mandates cap what can be conceded before a human is pulled in.
• Run records. Every workflow run keeps a per-step record you can open. Watchdog findings link to the evidence behind them.
• Plan limits. AI-marked agents need the org's plan to include AI, and runs count against your plan allowance (Plans, allowances & limits).

Troubleshooting

• It did not run. Check the agent's card in the Center: a paused status dot means it is switched off; a workflow with a weekly trigger runs about once a week, not at a fixed hour.
• It ran but the output is wrong. Open the run record and read the step where it went wrong; tighten the instructions on the workflow desk, or for analysts, edit the custom analyst's brief.
• Too noisy. Watchdog digests have per-page switches; workflows can gain an if then gate so only significant runs notify anyone.
• Nothing sends. A run showing Waiting for approval is paused on a human; the approvers were emailed and can also decide from the run page.