- APPROVED
- REJECTED
- PENDING
- DATA
Short answer
How can I use my AI to approve or reject proofs?
Your agent uses Ziflow tools to fetch asset metadata and then sends your final decision through your chat interface. You'll see how the data moves and where decisions land below.
The outcomes.
Where your decisions land.
The agent processes your instructions and updates the Ziflow status based on your input.
APPROVED
Final approval sent
The agent pushes an approval decision to Ziflow. This moves the asset forward in your production pipeline.
REJECTED
Rejection logged
The agent submits a rejection decision. This stops the asset from moving forward until you make changes.
PENDING
Human review needed
If the metadata is unclear, the agent flags the asset for you. You'll need to step in and decide manually.
DATA
Metadata retrieved
The agent pulls specific asset details from Ziflow. You get the context needed to make a call.
The workflow
What your AI does when the asset arrives.
The agent acts as the bridge between your chat and the Ziflow proofing environment.
Fetch asset details
The agent looks up the specific proof you are discussing to see its current status and metadata.
get_proofAnalyze metadata
Your agent reads the technical details and proofing information to prepare for your decision.
get_proofSubmit decision
Once you give the word, the agent sends the approval or rejection directly to the Ziflow platform.
submit_decisionConfirm action
The agent verifies the decision was successfully recorded in the system.
submit_decision
Try it
Copy these to start.
Use these prompts to trigger the Ziflow tools in your chat.
Starting points
These are opening lines. Swap the bracketed text with your actual Ziflow proof IDs or names.
Ziflow Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_HrMzZmZF6JgOlCK7FYpOKp1uLVLgEMgyzJucoQ8S/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Ziflow capabilities are ready to use.
{
"mcpServers": {
"ziflow-mcp": {
"url": "https://edge.vinkius.com/vk_preview_HrMzZmZF6JgOlCK7FYpOKp1uLVLgEMgyzJucoQ8S/mcp"
}
}
}Get the metadata for the proof with ID [ID_NUMBER] so I can review it.
Check the status of the [ASSET_NAME] proof in Ziflow.
Approve the proof for [ASSET_NAME] now.
Reject the [ASSET_NAME] proof because the colors are wrong.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Ziflow once, then ask.
Click the link to connect your Ziflow account. Your credentials stay encrypted on our side, and you can start chatting with your agent immediately.
Connect Ziflow to your AIFAQ
How the Ziflow task behaves.
- 01
Can the AI change the actual image or video file?
No. The agent only interacts with metadata and decision statuses. It cannot edit the content of the assets.
- 02
Does the AI decide if a proof is good or bad?
No. The agent provides the data and executes the command, but the final decision always comes from you.
- 03
Can I use this to upload new proofs?
No. This connection is only for retrieving existing proof data and submitting decisions.
- 04
Will the agent delete proofs from Ziflow?
No. There is no tool available to delete assets or proofs.
- 05
How does the agent know which proof I am talking about?
You must provide the proof ID or a specific name so the agent can find the right asset.
More questions about Ziflow? The Connector page answers them. See everything the Ziflow Connector can do
Connect Ziflow to Claude, Cursor, ChatGPT & more
