AI agents call get_collateral_info to retrieve information from Bybit without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves static or quasi-static reference data about collateral coins and their haircut rates. It has no side effects, does not modify account state, does not execute trades, and does not affect financial positions. It is a simple lookup operation that an AI agent could safely call to inform decision-making about collateral without risk of unintended consequences.
From the tool's definition Tool name and description indicate retrieval of collateral coin information (haircut rates, etc.) with no modification or execution capability. The verb 'Get' and the informational nature of 'collateral coin info' clearly denote a data retrieval operation.
Attacks that exploit this kind of access
Get collateral coin info (haircut rates, etc.). It is categorised as a Read tool in the Bybit MCP Server, which means it retrieves data without modifying state.
Register the Bybit MCP server in PolicyLayer and add a rule for get_collateral_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 Bybit. Nothing to install.
get_collateral_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_collateral_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_collateral_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_collateral_info is provided by the Bybit MCP server (johnnywic/bybit-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 →