// verdict · checked 2026-08-11 · Zapier Agents (2026 docs)
Zapier Agents
Zapier
Zaps that took a philosophy class. Still zaps.
- CAPABILITY
- 4/5 AGENT
- PROOF
- OPERATOR INSPECTABLEYou can run it and read the steps yourself — the record exists for whoever runs it
- TRACE ACCESS
- OPERATOR-OWNEDWhoever runs it holds the record; nothing is published
- LEASH
- Not measuredno execution record scored — measure it yourself
| LOOPSruns until doneyes | Docs cap Activities per run (10 Free/40 Pro) as a safeguard because step count varies with the task — one email costs 4 activities, two cost 7; a ceiling does not disqualify. |
| CHOOSESpicks its own toolsyes | The agent picks which connected action to call at runtime — docs say it "can still use any actions you've added, even if they do not appear in the instructions" — and infers field values itself. |
| ACTSchanges the worldyes | Actions run through real OAuth app connections; the "Take action" tab lists actions that modify data (send email, create records), each consuming an activity in the log. |
| RECOVERSfixes its own errorsno | On error the run goes to Needs action, which "requires your input to continue running"; the only alternative path is exception instructions the builder writes, not the agent re-deciding. |
| UNSUPERVISEDnobody watchingyes | Publishing an agent with a Schedule by Zapier or app-event trigger lets it run and complete tasks automatically; no documented mode requires per-action confirmation. |
Capability decisions combine documented behaviour and execution evidence. The proof label on each row shows which of the two decided it, and Leash is reported separately, only when it was measured from an execution record. Primary source: help.zapier.com. Tested against: Zapier Agents (2026 docs). Our confidence in this dossier: 8/10.
Can you see it work?
the weakest of the five grades aboveOPERATOR INSPECTABLE — You can run it and read the steps yourself — the record exists for whoever runs it. Trace access: OPERATOR-OWNED — Whoever runs it holds the record; nothing is published.
activity rows expand into run details: what it used and how it responded
Where we looked: Zapier help-center search for agent activity/history, then 'Review your agent's activity': run details show what the agent used and how it carried out instructions. · see the evidence →
- 2026-08-113/5 WRAPPER WITH AMBITION → 4/5 AGENTPre-launch adjudication: every criterion the verdict depends on was re-checked by hand against the published rules.editorial correction — no case was filed
Badges
earned, never for sale
These render from the live dataset, so a badge cannot outlive the verdict it claims. If the verdict changes the badge changes with it, and an unearned one returns 409 rather than an image. No sponsor can buy one, at any price.
Think this is wrong?
bring evidence, not opinionName the criterion and link a trace, a doc, or a recorded run that shows the behaviour. A verdict changed by evidence is the best thing that can happen to this site, and the change goes on the public record with your reason attached. Vendors are welcome, and a vendor's own filing is labelled as such rather than buried.