仅允许 SELECT/SHOW/DESCRIBE/EXPLAIN 语句
AI agents call dm8_execute_query to retrieve information from Mcp Dm8 without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool explicitly restricts execution to read-only SQL statements (SELECT, SHOW, DESCRIBE, EXPLAIN), meaning no data modification is possible. However, since it executes arbitrary SELECT queries, it could expose sensitive data across the entire database, warranting a medium severity. The tool name 'execute_query' combined with the read-only restriction keeps it in the Read category.
From the tool's definition 仅允许 SELECT/SHOW/DESCRIBE/EXPLAIN 语句 (Only allows SELECT/SHOW/DESCRIBE/EXPLAIN statements); server description says 'read-only query capabilities'
Documented attack patterns abuse exactly the kind of access dm8_execute_query gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Mcp Dm8, and nothing reaches the server without passing your rules. This is the rule we recommend for dm8_execute_query:
{
"version": "1",
"default": "deny",
"tools": {
"dm8_execute_query": {}
}
}dm8_execute_query is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
仅允许 SELECT/SHOW/DESCRIBE/EXPLAIN 语句. It is categorised as a Read tool in the Mcp Dm8 MCP Server, which means it retrieves data without modifying state.
Register the Mcp Dm8 MCP server in PolicyLayer and add a rule for dm8_execute_query: 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 Mcp Dm8. Nothing to install.
dm8_execute_query 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 dm8_execute_query 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 dm8_execute_query. 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.
dm8_execute_query is provided by the Mcp Dm8 MCP server (lianekai/mcp-dm8-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Mcp Dm8, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
3 Mcp Dm8 tools catalogued and risk-classified — across an index of 43,000+ MCP servers.