AI agents invoke run_ml_project to trigger actions in SAS MCP Server. 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 triggers execution of machine learning pipeline automation whose effects depend on the pipeline configuration and data inputs. While not inherently destructive, it runs complex automated operations on the SAS Viya platform that could consume resources, modify models, or produce unintended outputs if misconfigured.
From the tool's definition Tool executes an AutoML pipeline automation project, which runs code and operations on a SAS Viya compute environment.
Documented attack patterns abuse exactly the kind of access run_ml_project gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and SAS MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for run_ml_project:
{
"version": "1",
"default": "deny",
"tools": {
"run_ml_project": {
"limits": [
{
"counter": "run_ml_project_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}run_ml_project 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.
Run an AutoML pipeline automation project. It is categorised as a Execute tool in the SAS MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the SAS MCP Server MCP server in PolicyLayer and add a rule for run_ml_project: 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 SAS MCP Server. Nothing to install.
run_ml_project 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_ml_project 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_ml_project. 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_ml_project is provided by the SAS MCP Server MCP server (sassoftware/sas-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from SAS MCP Server, 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.
27 SAS MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.