AI agents call get_feed 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 retrieves and queries market intelligence data (feed items, tweets, signals) without creating, modifying, deleting, or executing code. It is purely informational/read-only in nature. The low severity reflects that misuse would only expose market data rather than cause financial loss, data destruction, or code execution. High confidence due to clear retrieval semantics in the description.
From the tool's definition Tool description explicitly states 'Get recent analyzed feed items, tweets, and breaking signals' - retrieval of existing data with no modification or side effects.
Attacks that exploit this kind of access
Get recent analyzed feed items, tweets, and breaking signals from Musashi. Use this when the user asks about recent news, social signals, or what is being discussed in prediction market circles. 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_feed: 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_feed 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_feed 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_feed. 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_feed 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 →