Query cloud infrastructure, SaaS, APIs, code and more with SQL. Queries are read-only and must use PostgreSQL syntax. For best performance: limit columns requested, use materialized CTEs instead of joins. Trust the search path unless sure you need to specify a schema. Check available tables and c...
AI agents call steampipe_query to retrieve information from Steampipe without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool only reads and queries data using PostgreSQL syntax and returns results. The description explicitly guarantees read-only semantics, preventing any modifications, deletions, or side effects. While it can query sensitive cloud infrastructure data, the read-only nature means misuse results in information disclosure rather than data loss or unauthorized changes.
From the tool's definition Tool description explicitly states 'Queries are read-only' and the tool is designed to 'Query cloud infrastructure, SaaS, APIs, code and more with SQL.' It retrieves data without modifying, deleting, or executing operations.
Documented attack patterns abuse exactly the kind of access steampipe_query gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Steampipe, and nothing reaches the server without passing your rules. This is the rule we recommend for steampipe_query:
{
"version": "1",
"default": "deny",
"tools": {
"steampipe_query": {}
}
}steampipe_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.
Query cloud infrastructure, SaaS, APIs, code and more with SQL. Queries are read-only and must use PostgreSQL syntax. For best performance: limit columns requested, use materialized CTEs instead of joins. Trust the search path unless sure you need to specify a schema. Check available tables and columns before querying using steampipe_table_list and steampipe_table_show. It is categorised as a Read tool in the Steampipe MCP Server, which means it retrieves data without modifying state.
Register the Steampipe MCP server in PolicyLayer and add a rule for steampipe_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 Steampipe. Nothing to install.
steampipe_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 steampipe_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 steampipe_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.
steampipe_query is provided by the Steampipe MCP server (turbot/steampipe-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 5 Steampipe tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
5 Steampipe tools catalogued and risk-classified — across an index of 42,500+ MCP servers.