count_weave_traces_tool
AI agents call count_weave_traces_tool to retrieve information from Weights & Biases MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The 'count' operation retrieves or aggregates trace data without modifying, deleting, or executing code. This is a read-only query operation with minimal blast radius—an AI misusing it would only retrieve counts of existing traces, not alter or delete data.
From the tool's definition Tool name contains 'count' which indicates a read operation. Server description specifies 'querying W&B Weave traces' as a capability, positioning this as a data retrieval function. No description provided for the tool itself, which limits certainty.
Attacks that exploit this kind of access
count_weave_traces_tool. It is categorised as a Read tool in the Weights & Biases MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Weights & Biases MCP Server MCP server in PolicyLayer and add a rule for count_weave_traces_tool: 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 Weights & Biases MCP Server. Nothing to install.
count_weave_traces_tool 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 count_weave_traces_tool 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 count_weave_traces_tool. 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.
count_weave_traces_tool is provided by the Weights & Biases MCP Server MCP server (riballes/mcp-server). 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 →