使用 DoWhy 进行 Rosenbaum 界限敏感性分析
AI agents call rosenbaum_bounds_analyzer to retrieve information from DoWhy MCP v2 0 without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Rosenbaum bounds sensitivity analysis is a statistical read/query operation that evaluates how sensitive causal inference results are to hidden confounders. It performs computation on existing data and returns analytical results without side effects, modifications, or code execution.
From the tool's definition 'Rosenbaum 界限敏感性分析' — Rosenbaum bounds is a sensitivity analysis technique that computes statistical bounds/estimates on causal effect estimates; it reads and analyzes existing data without modifying or executing anything
Attacks that exploit this kind of access
使用 DoWhy 进行 Rosenbaum 界限敏感性分析. It is categorised as a Read tool in the DoWhy MCP v2 0 MCP Server, which means it retrieves data without modifying state.
Register the DoWhy MCP v2 0 MCP server in PolicyLayer and add a rule for rosenbaum_bounds_analyzer: 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 DoWhy MCP v2 0. Nothing to install.
rosenbaum_bounds_analyzer 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 rosenbaum_bounds_analyzer 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 rosenbaum_bounds_analyzer. 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.
rosenbaum_bounds_analyzer is provided by the DoWhy MCP v2 0 MCP server (lesong36/dowhy_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 →