compute_returns
AI agents call compute_returns to retrieve information from Sablier MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Although the description is empty (lowering confidence), the context strongly suggests this computes returns from portfolio or market data for analysis. Computing/analyzing returns is a read operation with no side effects — it retrieves and processes data.
From the tool's definition Tool name is 'compute_returns' with no description provided. Based on sibling tools like 'analyze_derivatives', 'analyze_quantitative', and 'backtest_rules', this appears to be an analytical/computational tool that retrieves or calculates financial metrics…
Attacks that exploit this kind of access
compute_returns. It is categorised as a Read tool in the Sablier MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Sablier MCP Server MCP server in PolicyLayer and add a rule for compute_returns: 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 Sablier MCP Server. Nothing to install.
compute_returns 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 compute_returns 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 compute_returns. 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.
compute_returns is provided by the Sablier MCP Server MCP server (sablier-ai/sablier-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 →