AI agents use update_product_photo to create or update resources in Keycrm — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Keycrm environment.
This tool creates or modifies product data by updating the photo field. It is reversible (a new photo can be uploaded to replace the modified one), making it Write rather than Destructive. The blast radius is moderate: an AI agent could deface product listings or inject misleading imagery, affecting sales and customer trust, but the action is undoable.
From the tool's definition Tool description states 'Replace or add a photo on an existing product from a public URL' — modifies product data (photo) without permanent deletion or reversal constraints typical of Destructive actions.
Attacks that exploit this kind of access
Replace or add a photo on an existing product from a public URL. It is categorised as a Write tool in the Keycrm MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Keycrm MCP server in PolicyLayer and add a rule for update_product_photo: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Keycrm. Nothing to install.
update_product_photo is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the update_product_photo rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for update_product_photo. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
update_product_photo is provided by the Keycrm MCP server (ivanklymenko/keycrm-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →