promptthrift_suggest_model
AI agents call promptthrift_suggest_model to retrieve information from PromptThrift MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool appears to suggest or recommend a model based on context (likely from compressed conversation history or token counts). This is a read operation—it queries information and returns suggestions without side effects. The empty description lowers confidence slightly, but the name and sibling tools (compression, token counting) support a read/suggestion classification.
From the tool's definition Tool name 'promptthrift_suggest_model' indicates a suggestion/recommendation function. The server description mentions 'model routing' as a feature, suggesting this tool queries or recommends models.
Attacks that exploit this kind of access
promptthrift_suggest_model. It is categorised as a Read tool in the PromptThrift MCP MCP Server, which means it retrieves data without modifying state.
Register the PromptThrift MCP server in PolicyLayer and add a rule for promptthrift_suggest_model: 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 PromptThrift MCP. Nothing to install.
promptthrift_suggest_model 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 promptthrift_suggest_model 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 promptthrift_suggest_model. 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.
promptthrift_suggest_model is provided by the PromptThrift MCP server (woling-dev/promptthrift-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →