AI agents invoke docker to trigger actions in MCP SSH SRE. 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.
Docker operations cover a wide range of actions from read-only inspection to executing commands inside containers and destructive removal of containers/images/volumes. Given the ambiguity, Execute is the most appropriate base category given the blast radius of arbitrary Docker operations. Confidence is moderate due to the uninformative description.
From the tool's definition Tool name 'docker' and description 'Docker ops.' — description is uninformative but 'docker' operations can span reading container info, starting/stopping containers, executing commands inside containers, and potentially destructive operations like removing…
Documented attack patterns abuse exactly the kind of access docker gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MCP SSH SRE, and nothing reaches the server without passing your rules. This is the rule we recommend for docker:
{
"version": "1",
"default": "deny",
"tools": {
"docker": {
"limits": [
{
"counter": "docker_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}docker 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.
Docker ops. It is categorised as a Execute tool in the MCP SSH SRE MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the MCP SSH SRE MCP server in PolicyLayer and add a rule for docker: 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 MCP SSH SRE. Nothing to install.
docker 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 docker 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 docker. 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.
docker is provided by the MCP SSH SRE MCP server (jeprecated/mcp-ssh-sre). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from MCP SSH SRE, 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.
13 MCP SSH SRE tools catalogued and risk-classified — across an index of 43,000+ MCP servers.