Query the sportsbook +EV endpoint and return candidate plays for enabled books. Secondary discovery only — use /screen for primary playable-bet selection. Set validated=true to run sharp-movement validation on candidates. Returns 0 rows on quiet days when no +EV opportunities exist — that is norm...
AI agents call ev_candidates to retrieve information from PropProfessor MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool queries a sportsbook endpoint and returns candidate plays (+EV opportunities). This is a read/fetch operation returning data for analysis. No money is moved, no bets are placed — it is explicitly described as 'secondary discovery only' for playable-bet selection, not execution.
From the tool's definition 'Query the sportsbook +EV endpoint and return candidate plays' — retrieves/queries data with no side effects indicated
Attacks that exploit this kind of access
Query the sportsbook +EV endpoint and return candidate plays for enabled books. Secondary discovery only — use /screen for primary playable-bet selection. Set validated=true to run sharp-movement validation on candidates. Returns 0 rows on quiet days when no +EV opportunities exist — that is normal, not a bug. It is categorised as a Read tool in the PropProfessor MCP MCP Server, which means it retrieves data without modifying state.
Register the PropProfessor MCP server in PolicyLayer and add a rule for ev_candidates: 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 PropProfessor MCP. Nothing to install.
ev_candidates 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 ev_candidates 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 ev_candidates. 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.
ev_candidates is provided by the PropProfessor MCP server (j17drake/propprofessor-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 →