Substructure search by SMILES
AI agents call search_substructure to retrieve information from Chemspace MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves chemical data based on molecular structure queries expressed as SMILES strings. It has no capability to create, modify, delete, or execute code. Even if an AI agent misuses it by querying unintended compounds, the worst outcome is information disclosure of existing chemical data, which is a low-severity read operation.
From the tool's definition Tool is described as performing a 'substructure search' which queries the Chemspace API for chemical compounds without modifying data.
Attacks that exploit this kind of access
Substructure search by SMILES. It is categorised as a Read tool in the Chemspace MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Chemspace MCP Server MCP server in PolicyLayer and add a rule for search_substructure: 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 Chemspace MCP Server. Nothing to install.
search_substructure 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 search_substructure 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 search_substructure. 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.
search_substructure is provided by the Chemspace MCP Server MCP server (lukasmki/chemspace-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 →