使用 DoWhy 进行异常归因分析
AI agents call anomaly_attribution_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.
The tool performs attribution analysis (identifying root causes of anomalies) using DoWhy, which is a read/query-style analytical operation — it computes and returns attribution scores without modifying data. The sibling tools like 'attribute_anomalies_gcm' and 'anomaly_scores_gcm' confirm this is an analytical pattern.
From the tool's definition '异常归因分析' translates to 'anomaly attribution analysis' — this is an analytical/read operation that queries and attributes anomalies using DoWhy causal inference
Attacks that exploit this kind of access
使用 DoWhy 进行异常归因分析. 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 anomaly_attribution_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.
anomaly_attribution_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 anomaly_attribution_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 anomaly_attribution_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.
anomaly_attribution_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 →