Execute Anonymous Apex
AI agents invoke anonymous-apex to trigger actions in Salesforce MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool allows execution of arbitrary code (Apex scripts) in the Salesforce environment. While not immediately destructive on its own, execution of untrusted or misconfigured Apex code could modify data, trigger unwanted business logic, access sensitive information, or interact with external systems.
From the tool's definition Tool name 'anonymous-apex' and description 'Execute Anonymous Apex' indicate execution of arbitrary Apex code within Salesforce. The description explicitly uses the verb 'Execute' combined with 'Anonymous' (unrestricted code).
Attacks that exploit this kind of access
Execute Anonymous Apex. It is categorised as a Execute tool in the Salesforce MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Salesforce MCP server in PolicyLayer and add a rule for anonymous-apex: 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 Salesforce MCP. Nothing to install.
anonymous-apex is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the anonymous-apex 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 anonymous-apex. 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.
anonymous-apex is provided by the Salesforce MCP server (jpmonette/salesforce-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 →