Load the current agent state from the state file.
AI agents call agent_state_load_state to retrieve information from Agent State 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 or queries existing state data from persistent storage. It performs no creation, modification, deletion, or execution of external operations. The verb 'load' is passive and indicates data retrieval only. As a read operation with minimal blast radius if misused by an AI agent, it is classified as Read with low severity.
From the tool's definition Tool name 'agent_state_load_state' and description 'Load the current agent state from the state file' indicate a read-only retrieval operation with no modification or side effects.
Attacks that exploit this kind of access
Load the current agent state from the state file. It is categorised as a Read tool in the Agent State MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Agent State MCP Server MCP server in PolicyLayer and add a rule for agent_state_load_state: 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 Agent State MCP Server. Nothing to install.
agent_state_load_state 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 agent_state_load_state 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 agent_state_load_state. 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.
agent_state_load_state is provided by the Agent State MCP Server MCP server (tianhuil/agent-state). 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 →