AI agents call os_inspect_at to retrieve information from OScribe without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries the current state of a UI element without modifying, executing, or deleting anything. It is a read-only inspection operation analogous to taking a screenshot or querying element metadata. While it operates within a desktop automation context, the tool itself performs no actions that change system state, execute code, or commit resources.
From the tool's definition Tool description states it 'Get[s] the UI element at specific coordinates' and 'Returns element type, name, bounds, and state' — purely retrieval operations with no side effects.
Attacks that exploit this kind of access
Get the UI element at specific coordinates. Returns element type, name, bounds, and state. It is categorised as a Read tool in the OScribe MCP Server, which means it retrieves data without modifying state.
Register the OScribe MCP server in PolicyLayer and add a rule for os_inspect_at: 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 OScribe. Nothing to install.
os_inspect_at is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the os_inspect_at 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 os_inspect_at. 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.
os_inspect_at is provided by the OScribe MCP server (mikealkeal/oscribe). 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 →