Database analysis. Actions: schema|table_usage|dead_tables
AI agents call db to retrieve information from Project Graph without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The db tool performs introspective queries on database structure and usage patterns. While it could inform decisions about destructive actions, the tool itself only reads data. However, severity is medium rather than low because database schema and table usage information could be sensitive in some contexts and could inform reconnaissance for attacks.
From the tool's definition Tool description states 'Database analysis' with actions limited to 'schema|table_usage|dead_tables' — all read-only query operations that retrieve metadata and usage information without modification or deletion.
Attacks that exploit this kind of access
Database analysis. Actions: schema|table_usage|dead_tables. It is categorised as a Read tool in the Project Graph MCP Server, which means it retrieves data without modifying state.
Register the Project Graph MCP server in PolicyLayer and add a rule for db: 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 Project Graph. Nothing to install.
db 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 db 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 db. 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.
db is provided by the Project Graph MCP server (rnd-pro/project-graph-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 →