AI agents call get_arbitrage to retrieve information from Musashi without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool reads and analyzes market data to identify price discrepancies across platforms. It has no side effects, does not execute trades, does not modify data, and does not commit financial transactions. While it informs financial decision-making, the tool itself only retrieves and presents information.
From the tool's definition Tool description states 'Find current arbitrage opportunities' - queries and retrieves market data showing price discrepancies and trading opportunities between platforms. No modification, deletion, or execution of trades occurs.
Attacks that exploit this kind of access
Find current arbitrage opportunities between Polymarket and Kalshi. Use this when the user asks about price discrepancies, spreads, or cross-platform trading opportunities. It is categorised as a Read tool in the Musashi MCP Server, which means it retrieves data without modifying state.
Register the Musashi MCP server in PolicyLayer and add a rule for get_arbitrage: 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 Musashi. Nothing to install.
get_arbitrage 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_arbitrage 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_arbitrage. 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_arbitrage is provided by the Musashi MCP server (musashibot/musashi-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 →