Get the most recent failed calls with full error details and retry logs.
AI agents call recent_errors to retrieve information from SelfHeal MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves historical error and retry information for monitoring and debugging purposes. It performs no data modification, deletion, code execution, or financial operations. The blast radius of misuse is minimal — an attacker could only access observability logs to understand system behavior, not affect actual functionality or data integrity.
From the tool's definition Tool name is 'recent_errors' and description states 'Get the most recent failed calls with full error details and retry logs' — uses read-only verb 'Get' and returns observability/logging data with no modification or side effects.
Attacks that exploit this kind of access
Get the most recent failed calls with full error details and retry logs. It is categorised as a Read tool in the SelfHeal MCP MCP Server, which means it retrieves data without modifying state.
Register the SelfHeal MCP server in PolicyLayer and add a rule for recent_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 SelfHeal MCP. Nothing to install.
recent_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 recent_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 recent_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.
recent_errors is provided by the SelfHeal MCP server (selfheal-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 →