This record as markdown: /tools/openclaw/models.md
What models does on OpenClaw
AI agents call models to retrieve information from OpenClaw without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Why models is rated Low
The description includes both Read ('List, scan') and Write ('set') operations. Per the rules, the most severe applicable category should be chosen, making this a Write. However, 'set model providers' is a configuration change with limited blast radius. Given the ambiguity and that 'list/scan' dominate the description, I lean toward Write but with moderate confidence.
From the tool's definition 'List, scan, and set model providers' — the tool lists and scans providers (Read operations), but also includes 'set' which implies a Write operation.
Attacks that exploit this kind of access
The rule that runs models safely
PolicyLayer is an MCP gateway: it sits between your AI agents and OpenClaw, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For models, this is the rule to start with:
models is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect OpenClaw, apply this rule, and every models call is checked against it from then on.
Questions about models
List, scan, and set model providers. It is categorised as a Read tool in the OpenClaw MCP Server, which means it retrieves data without modifying state.
Register the OpenClaw MCP server in PolicyLayer and add a rule for models: 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 OpenClaw. Nothing to install.
models is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the models 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 models. 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.
models is provided by the OpenClaw MCP server (openclaw). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on OpenClaw, and thousands of servers like it.
This server
Across the catalogue