AI agents invoke run_tests to trigger actions in Salesforce. 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 runs Apex unit tests in Salesforce, which is a code execution action that can modify system state (create test data, update records, log results, trigger callbacks). While tests are typically designed to be isolated, they execute arbitrary Apex code in the Salesforce org and could have side effects.
From the tool's definition Tool name 'run_tests' with description 'Execute Apex tests' indicates execution of code (Apex tests) in the Salesforce environment. This is a code execution operation that triggers external operations whose effects depend on test implementation.
Documented attack patterns abuse exactly the kind of access run_tests gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Salesforce, and nothing reaches the server without passing your rules. This is the rule we recommend for run_tests:
{
"version": "1",
"default": "deny",
"tools": {
"run_tests": {
"limits": [
{
"counter": "run_tests_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}run_tests stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Execute Apex tests. It is categorised as a Execute tool in the Salesforce 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 run_tests: 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. Nothing to install.
run_tests 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 run_tests 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 run_tests. 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.
run_tests is provided by the Salesforce MCP server (ryu-727/salesforce-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Salesforce, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
25 Salesforce tools catalogued and risk-classified — across an index of 43,000+ MCP servers.