Search a log group for a pattern (default ERROR) in an absolute or relative window. Returns timestamp + instance + message.
AI agents call search_errors to retrieve information from Cloudwatch without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries log data without any side effects. It performs a search operation that returns existing log entries matching a pattern—a classic Read operation. The read-only nature of the CloudWatch Logs API and explicit server-level read-only constraint confirm this classification.
From the tool's definition Tool description states it 'Search[es] a log group for a pattern' and 'Returns timestamp + instance + message'. Server description emphasizes 'read-only access' and 'log analysis'. No modification, deletion, or execution capability is indicated.
Attacks that exploit this kind of access
Search a log group for a pattern (default ERROR) in an absolute or relative window. Returns timestamp + instance + message. It is categorised as a Read tool in the Cloudwatch MCP Server, which means it retrieves data without modifying state.
Register the Cloudwatch MCP server in PolicyLayer and add a rule for search_errors: 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 Cloudwatch. Nothing to install.
search_errors 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 search_errors 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 search_errors. 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.
search_errors is provided by the Cloudwatch MCP server (shivam-singh-au17/cloudwatch-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 →