AI agents call get_diagram_info to retrieve information from Arch7 without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The 'get_' prefix conventionally indicates a read/fetch operation that retrieves information without modification. In the context of a diagram tool, retrieving diagram metadata or structure is a safe query with no side effects. The empty description prevents complete certainty, but the name alone is sufficient to classify as Read with high confidence. Low severity: worst case is exposure of benign diagram metadata.
From the tool's definition Tool name 'get_diagram_info' combined with its sibling context (create, mermaid_to_excalidraw, modify) strongly suggests a retrieval operation. Lack of description reduces confidence slightly, but 'get_' prefix is a standard Read operation indicator.
Attacks that exploit this kind of access
get_diagram_info. It is categorised as a Read tool in the Arch7 MCP Server, which means it retrieves data without modifying state.
Register the Arch7 MCP server in PolicyLayer and add a rule for get_diagram_info: 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 Arch7. Nothing to install.
get_diagram_info 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 get_diagram_info 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 get_diagram_info. 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.
get_diagram_info is provided by the Arch7 MCP server (ktcrisis/arch7). 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 →