Generate training configuration for synthetic data models. Returns a cached train_config_id for use in start_training_workflow. Supports TabGAN (rf_tab_gan) and TimeGAN (rf_time_gan).
AI agents call obtain_train_config to retrieve information from Rockfish MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool produces a read-only output (a configuration ID) that is passed to other tools like start_training_workflow. It does not execute training, modify databases, or trigger external operations. The action is purely informational/generative with no side effects. Even though it supports different model types (TabGAN, TimeGAN), these are parameters to the configuration generation, not execution triggers.
From the tool's definition Tool description states it 'Generate[s] training configuration' and 'Returns a cached train_config_id' — it retrieves or generates a configuration object for later use without modifying state. No data is created, deleted, or irreversibly changed.
Attacks that exploit this kind of access
Generate training configuration for synthetic data models. Returns a cached train_config_id for use in start_training_workflow. Supports TabGAN (rf_tab_gan) and TimeGAN (rf_time_gan). It is categorised as a Read tool in the Rockfish MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Rockfish MCP Server MCP server in PolicyLayer and add a rule for obtain_train_config: 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 Rockfish MCP Server. Nothing to install.
obtain_train_config 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 obtain_train_config 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 obtain_train_config. 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.
obtain_train_config is provided by the Rockfish MCP Server MCP server (wolfdancer/rockfish-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 →