Build and launch a Bazel macOS application locally via
AI agents invoke bazel_macos_run to trigger actions in XcodeBazelMCP. 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 executes code by building and launching macOS applications, which is a code execution action. The blast radius is high because a malicious agent could build and run arbitrary applications with the privileges of the user running the tool, potentially leading to system compromise, data exfiltration, or other harmful side effects.
From the tool's definition Tool name 'bazel_macos_run' and description 'Build and launch a Bazel macOS application locally' indicate the tool builds and executes arbitrary macOS applications.
Documented attack patterns abuse exactly the kind of access bazel_macos_run gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and XcodeBazelMCP, and nothing reaches the server without passing your rules. This is the rule we recommend for bazel_macos_run:
{
"version": "1",
"default": "deny",
"tools": {
"bazel_macos_run": {
"limits": [
{
"counter": "bazel_macos_run_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}bazel_macos_run 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.
Build and launch a Bazel macOS application locally via. It is categorised as a Execute tool in the XcodeBazelMCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the XcodeBazel MCP server in PolicyLayer and add a rule for bazel_macos_run: 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 XcodeBazelMCP. Nothing to install.
bazel_macos_run 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 bazel_macos_run 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 bazel_macos_run. 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.
bazel_macos_run is provided by the XcodeBazel MCP server (xcodebazelmcp/xcodebazelmcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from XcodeBazelMCP, 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.
117 XcodeBazelMCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.