Get ingestion statistics for a stream (event count, rate, schema version)
AI agents call conduit_stream_stats to retrieve information from Conduit MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool only retrieves monitoring/observability data about stream performance metrics. It does not ingest, modify, delete, or execute operations—purely informational. The blast radius of misuse is minimal; an agent querying statistics cannot harm data or systems.
From the tool's definition Tool retrieves ingestion statistics for a stream (event count, rate, schema version) with no modification or destructive operations. The description uses query-only language: 'Get' indicates data retrieval without side effects.
Attacks that exploit this kind of access
Get ingestion statistics for a stream (event count, rate, schema version). It is categorised as a Read tool in the Conduit MCP MCP Server, which means it retrieves data without modifying state.
Register the Conduit MCP server in PolicyLayer and add a rule for conduit_stream_stats: 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 Conduit MCP. Nothing to install.
conduit_stream_stats 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 conduit_stream_stats 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 conduit_stream_stats. 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.
conduit_stream_stats is provided by the Conduit MCP server (useconduit/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 →