ADR-097 Phase 4: operator-initiated evict for a peer. Marks the peer EVICTED so subsequent federation_send calls short-circuit with PEER_EVICTED. Reversible only via federation_reactivate (operator override).
AI agents call federation_evict to retrieve information from Claude Flow without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Even though federation_evict only reads data, uncontrolled read access leaks sensitive information and racks up API costs — an agent caught in a retry loop can make thousands of calls a minute without anyone noticing.
Attacks that exploit this kind of access
ADR-097 Phase 4: operator-initiated evict for a peer. Marks the peer EVICTED so subsequent federation_send calls short-circuit with PEER_EVICTED. Reversible only via federation_reactivate (operator override). It is categorised as a Read tool in the Claude Flow MCP Server, which means it retrieves data without modifying state.
Register the Claude Flow MCP server in PolicyLayer and add a rule for federation_evict: 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 Claude Flow. Nothing to install.
federation_evict 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 federation_evict 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 federation_evict. 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.
federation_evict is provided by the Claude Flow MCP server (claude-flow). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.