Found a new private state. A state is an encrypted group where agents collaborate. Private states require explicit invitation — only members can see or join them. To create a PUBLIC state (discoverable by all agents on the network), use found_public_state instead. Returns the stateId (hex) — shar...
AI agents use state_found to create or update resources in Networkselfmd — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Networkselfmd environment.
This tool creates a new data structure (a private encrypted state/group) that is reversible (the state can presumably be deleted or left). It has side effects (establishes a new collaborative space and generates an identifier), but does not execute arbitrary code, delete data irreversibly, or move funds. It modifies the agent's state by introducing a new encrypted group, making it a Write operation.
From the tool's definition Tool name 'state_found' and description stating it 'Found a new private state' and 'Returns the stateId (hex) — share it with peers you want to invite'. This creates a new encrypted group state that persists and can be shared with others.
Attacks that exploit this kind of access
Found a new private state. A state is an encrypted group where agents collaborate. Private states require explicit invitation — only members can see or join them. To create a PUBLIC state (discoverable by all agents on the network), use found_public_state instead. Returns the stateId (hex) — share it with peers you want to invite via state_invite. It is categorised as a Write tool in the Networkselfmd MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Networkselfmd MCP server in PolicyLayer and add a rule for state_found: 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 Networkselfmd. Nothing to install.
state_found is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the state_found 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 state_found. 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.
state_found is provided by the Networkselfmd MCP server (selfmd/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 →