查询已学习的知识，包括技能、因果模式、语义概念。
AI agents call screen_query_knowledge to retrieve information from Screen Agent without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a Read operation - it retrieves stored knowledge without side effects. However, severity is elevated to 'high' because the queried knowledge base (learned UI interactions, causal patterns, semantic understanding) could reveal sensitive information about system behavior, user workflows, application vulnerabilities, or training data that an adversarial agent could exploit to craft more effective attacks on the…
From the tool's definition Tool name 'screen_query_knowledge' and description indicate querying (reading) previously learned knowledge including skills, causal patterns, and semantic concepts. No modification or execution occurs.
Attacks that exploit this kind of access
查询已学习的知识，包括技能、因果模式、语义概念。. It is categorised as a Read tool in the Screen Agent MCP Server, which means it retrieves data without modifying state.
Register the Screen Agent MCP server in PolicyLayer and add a rule for screen_query_knowledge: 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 Screen Agent. Nothing to install.
screen_query_knowledge 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 screen_query_knowledge 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 screen_query_knowledge. 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.
screen_query_knowledge is provided by the Screen Agent MCP server (lqszhsp/screen-agent). 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 →