predict_stability
AI agents call predict_stability to retrieve information from SPIRED-Stab 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 retrieve stability predictions from a deep learning model. This is a read operation—it queries the model and returns results without creating, modifying, deleting, or executing arbitrary code. The lack of side effects (no write/delete operations, no financial transactions, no shell execution) places it in the Read category.
From the tool's definition Tool name 'predict_stability' combined with server description stating it 'Enables protein stability prediction ($DeltaDelta$G and $Delta$Tm)' indicates a query/inference operation.
Attacks that exploit this kind of access
predict_stability. It is categorised as a Read tool in the SPIRED-Stab MCP MCP Server, which means it retrieves data without modifying state.
Register the SPIRED-Stab MCP server in PolicyLayer and add a rule for predict_stability: 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 SPIRED-Stab MCP. Nothing to install.
predict_stability 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 predict_stability 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 predict_stability. 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.
predict_stability is provided by the SPIRED-Stab MCP server (macromnex/spired_stab_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 →