AI agents call get_results to retrieve information from Fairchem without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and retrieves pre-computed simulation results. It has no side effects, does not modify state, does not execute new operations, and does not delete or overwrite data. The blast radius of misuse is minimal—an agent could only read results it should not see, which is a confidentiality concern rather than a security or operational risk. Typical of Read category tools like 'get', 'fetch', or 'query'.
From the tool's definition Tool description states 'Return a job's final results' — a retrieval operation with no modification. The results retrieved (NEB barrier/energies, phonon gamma) are computed outputs, not data being written or deleted.
Attacks that exploit this kind of access
Return a job's final results, if any: NEB barrier/energies, phonon gamma. It is categorised as a Read tool in the Fairchem MCP Server, which means it retrieves data without modifying state.
Register the Fairchem MCP server in PolicyLayer and add a rule for get_results: 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 Fairchem. Nothing to install.
get_results 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 get_results 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 get_results. 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.
get_results is provided by the Fairchem MCP server (jkitchin/fairchem-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 →