Ziflow
ProductivityConnect AI agents to Ziflow creative proofing. Agents create and update proofs, invite reviewers, collect comments and review decisions, and read folders and users, so approval status flows into your project and messaging tools without manual chasing.
What This Integration Enables
Ziflow is where creative work goes to get a decision: a proof carries the asset, the reviewers, their per-stage decisions, and every annotation pinned to a specific page or frame. What most teams are missing is not the proofing tool but the connective tissue around it, and that is what this connector is for. FlowRunner agents create proofs the moment files are delivered, watch review progress without a human refreshing the page, and move decisions and comments into the systems where work actually gets scheduled. Proofing is already a human-judgment workflow; the point of automating around it is to protect that judgment from the logistics that bury it. - Open proofs automatically from cloud storage or design tool deliveries, reviewers pre-invited - Route decisions like approved and changes required into project and messaging tools - Convert comments and annotations into tickets and change logs - Keep a review-status audit trail in a spreadsheet without manual copying - Chase stalled reviews automatically, so the bottleneck reviewer hears from a bot, not a colleague One caveat stated plainly: Ziflow's public REST API is partially documented, so this connector is offered best-effort. Verify behavior against the live API before depending on it in production.
Without FlowRunner
With FlowRunner
Use Case Scenarios
The proof that opens before anyone asks
A final-render file lands in [Dropbox](/integrations/dropbox). The agent resolves the client's folder with List Folders, looks up the account team's reviewer emails with List Users, and calls Create Proof with the file URL. Ziflow ingests the file, reviewers get their invitations, and the proof link posts to [Slack](/integrations/slack) with the deadline attached. The producer never touches an upload form, and review starts the same hour the file is delivered instead of after the next standup.
Annotations become tickets, not memories
A reviewer marks changes required. The agent pulls every annotation with List Comments: the text, the author, the exact page and markup coordinates. Each one becomes an issue in [Jira](/integrations/jira) with the proof link and coordinates in the description, batched under the revision epic. The designer works a ticket list instead of scrolling a comment thread, and when the next version uploads, the team can point to exactly which requests it addressed.
Sign-off tracked to the minute, logged for the audit
For a regulated client, every approval needs a paper trail. On a schedule, the agent calls List Reviews on each active proof and writes the decision state to [Google Sheets](/integrations/google-sheets): reviewer, decision, stage, timestamp. When the last decision lands, the agent posts the sign-off summary and calls Update Proof to move the proof to its completed status. Months later, the question "who approved this and when" is answered by a spreadsheet, not by archaeology.
Human-in-Loop Highlight
Update Proof can set a proof's status to Approved, and in every downstream system that status is read as permission: permission to publish the campaign, send the files to the printer, push the video live. An agent that flips that field on its own initiative has effectively forged a sign-off, and a print run against a wrongly approved proof is money and time nobody recovers. So in FlowRunner workflows the approval status is earned, never asserted: the agent reads the actual per-reviewer decisions with List Reviews, and only when every required reviewer has personally recorded approved does Update Proof advance the status, with the decision log attached to the change. The agent handles the watching, collating, and chasing. The word "approved" only ever originates from a human reviewer inside the proof.
Agent Capabilities
8 actionsProofs
4- List Proofs Retrieves a paginated list of proofs in the account, each with identifier, name, status, and review progress. The scan behind status dashboards and stalled-review sweeps.
- Get Proof Retrieves full details for a single proof: name, current status, version, files, reviewers, and folder. Used to gather context before notifying or updating.
- Create Proof Creates a proof from one or more file URLs and optionally assigns reviewers, who are invited once the proof is ready. The action that turns a delivered file into a running review.
- Update Proof Updates a proof's editable properties such as name, status, or folder, changing only the fields you provide. Status moves to Approved or Completed run through this page's human gate.
Reviews
2- List Reviews Retrieves the decisions recorded for a specific proof: each reviewer's verdict, the review stage, and timestamps. The source of truth for sign-off state.
- List Comments Retrieves the comments and annotations on a specific proof, including text, author, file and page, and markup coordinates. The raw material for tickets and change logs.
Organization
2- List Folders Retrieves the folders that organize proofs, with ID, name, and parent. Used to file new proofs in the right place and filter reporting by client or project.
- List Users Retrieves the account's members and reviewers with ID, name, email, and role. Used to resolve reviewer emails when creating proofs and to audit membership.
Frequently Asked Questions
What can FlowRunner do with Ziflow?
FlowRunner agents can run List Proofs, Get Proof, and Create Proof in Ziflow, plus 5 more actions.
Does connecting Ziflow to FlowRunner require OAuth?
No. Ziflow connects to FlowRunner with an API key, no OAuth flow required.
Can Ziflow trigger a FlowRunner workflow automatically?
Ziflow doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.
Start building with Ziflow
$100 in credits. No card required. Connect in minutes.