AI agents call cap_entities to retrieve information from Sflight without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a metadata inspection tool that queries the structure of a CAP project to discover available entities. It performs read-only retrieval of schema information without creating, modifying, deleting, or executing operations. The blast radius of misuse is minimal—an agent could only learn what entities exist, not access or alter any actual data.
From the tool's definition Tool description states it 'List[s] all CDS entities' - a retrieval/enumeration operation with no modification capability. The data retrieved (entity metadata: key fields, element counts, associations) are schema/catalog information, not operational data.
Attacks that exploit this kind of access
List all CDS entities in the CAP project with their key fields, element count, and association/composition counts. It is categorised as a Read tool in the Sflight MCP Server, which means it retrieves data without modifying state.
Register the Sflight MCP server in PolicyLayer and add a rule for cap_entities: 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 Sflight. Nothing to install.
cap_entities 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 cap_entities 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 cap_entities. 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.
cap_entities is provided by the Sflight MCP server (mindsetconsulting/sflight-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 →