AI agents use resolve to create or update resources in Instruckt — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Instruckt environment.
The tool creates or modifies data reversibly by changing an annotation's resolved status. It is not destructive because the annotation data itself persists (only the marker display is affected), and the change can theoretically be reversed by marking it unresolved again. This is a Write-category operation.
From the tool's definition Tool description states it 'Mark an annotation as resolved', which modifies the state of an annotation record. This is a state-changing operation that alters persisted data (removing markers).
Attacks that exploit this kind of access
Mark an annotation as resolved. This removes the marker from the browser on next page load. It is categorised as a Write tool in the Instruckt MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Instruckt MCP server in PolicyLayer and add a rule for resolve: 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 Instruckt. Nothing to install.
resolve 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 resolve 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 resolve. 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.
resolve is provided by the Instruckt MCP server (tdwesten/instruckt-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 →