AI agents use install_library to create or update resources in Databricks MCP Server — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Databricks MCP Server environment.
Installing libraries modifies the cluster environment by adding software packages. This is a Write operation (reversible in principle, as libraries can be uninstalled), but carries high severity because it can introduce malicious or incompatible packages that affect all workloads running on the cluster, potentially compromising security or stability at scale.
From the tool's definition 'Install libraries on a cluster' — installs software packages onto a running cluster
Documented attack patterns abuse exactly the kind of access install_library gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Databricks MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for install_library:
{
"version": "1",
"default": "deny",
"tools": {
"install_library": {
"limits": [
{
"counter": "install_library_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}install_library stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Install libraries on a cluster. It is categorised as a Write tool in the Databricks MCP Server MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Databricks MCP Server MCP server in PolicyLayer and add a rule for install_library: 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 Databricks MCP Server. Nothing to install.
install_library is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the install_library 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 install_library. 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.
install_library is provided by the Databricks MCP Server MCP server (markov-kernel/databricks-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Databricks 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.
38 Databricks MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.