AI agents call get_table_stats to retrieve information from Omnibase without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and returns statistical metadata about table columns without modifying data or executing code. While it accesses database information, it is purely analytical and read-only. The sampling approach further confirms it performs passive analysis rather than side effects. Low severity because misuse would only expose statistical summaries, not sensitive row-level data directly.
From the tool's definition Tool performs data retrieval only: 'Column-level statistics for a table: null counts/percentages, distinct values, min/max.
Attacks that exploit this kind of access
Column-level statistics for a table: null counts/percentages, distinct values, min/max. Uses a sample (default 10,000 rows) to avoid full table scans on large databases. It is categorised as a Read tool in the Omnibase MCP Server, which means it retrieves data without modifying state.
Register the Omnibase MCP server in PolicyLayer and add a rule for get_table_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 Omnibase. Nothing to install.
get_table_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 get_table_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 get_table_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.
get_table_stats is provided by the Omnibase MCP server (omnibase-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 →