Unpublish a document (moves to drafts.*). Requires write token.
AI agents call sanity_unpublish to permanently remove resources in Sanity MCP Server — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
Unpublishing a document removes it from public availability, which is a significant irreversible-in-effect action even though the document technically moves to draft state. While the document data isn't deleted, the public content is taken down immediately, which can have broad impact on live sites. The 'Requires write token' confirms it causes real state changes.
From the tool's definition Unpublish a document (moves to drafts.*)
Attacks that exploit this kind of access
Unpublish a document (moves to drafts.*). Requires write token. It is categorised as a Destructive tool in the Sanity MCP Server MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Sanity MCP Server MCP server in PolicyLayer and add a rule for sanity_unpublish: 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 Sanity MCP Server. Nothing to install.
sanity_unpublish is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the sanity_unpublish 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 sanity_unpublish. 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.
sanity_unpublish is provided by the Sanity MCP Server MCP server (purple-horizons/sanity-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 →