cost_signals
AI agents call cost_signals to retrieve information from AWS MCP Audit without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Cost signal retrieval is fundamentally a Read operation—it queries AWS billing or cost data to provide insight without creating, modifying, deleting, or executing external operations. The audit-focused server context and read-only design reinforce this classification.
From the tool's definition Tool is exposed on an 'Audit' server described as 'read-only assessment' of AWS environments. Sibling tools (cost_explorer_by_region, cost_explorer_by_service, cost_explorer_summary_tool) indicate cost analysis capabilities.
Attacks that exploit this kind of access
cost_signals. It is categorised as a Read tool in the AWS MCP Audit MCP Server, which means it retrieves data without modifying state.
Register the AWS MCP Audit MCP server in PolicyLayer and add a rule for cost_signals: 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 AWS MCP Audit. Nothing to install.
cost_signals 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 cost_signals 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 cost_signals. 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.
cost_signals is provided by the AWS MCP Audit MCP server (oldcoder01/aws-mcp-audit). 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 →