Get system metrics and performance data Use when native Bash is wrong because you need Ruflo runtime metrics (HNSW index size, ReasoningBank state, swarm health, breaker status) — those are not in /proc, only in the running daemon. For OS-level info (uptime, disk, mem), native Bash + standard too...
AI agents call system_metrics 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.
This tool retrieves and queries runtime metrics and performance data without modifying, executing, or deleting anything. It is purely observational/diagnostic in nature, similar to native monitoring commands. The blast radius of misuse is minimal—returning incorrect metrics would not damage systems or trigger unintended operations. Classified as Read with low severity.
From the tool's definition Tool description states 'Get system metrics and performance data' with specific read-only examples: 'HNSW index size, ReasoningBank state, swarm health, breaker status'. All operations are queries returning monitoring information with no side effects.
Risk signalsAdmin/system-level operation
Attacks that exploit this kind of access
Get system metrics and performance data Use when native Bash is wrong because you need Ruflo runtime metrics (HNSW index size, ReasoningBank state, swarm health, breaker status) — those are not in /proc, only in the running daemon. For OS-level info (uptime, disk, mem), native Bash + standard tools are fine. 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 system_metrics: 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.
system_metrics 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 system_metrics 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 system_metrics. 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.
system_metrics 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.