// SCAN REPORT

Postgres

10 tools. 4 can modify or destroy data without limits.

1 destructive tool with no built-in limits. Policy required.

Last updated: 30 Jun 2026

4 can modify or destroy data

6 read-only

10 tools total

Community server · catalogue entry verified 30/06/2026

How to control Postgres ↓

TOOL MAP

What Postgres exposes to your agents

Read (6) Write / Execute (3) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Postgres tools

Destructive · High delete_rows DELETE operations are irreversible and permanently remove data. Write · High insert_row INSERT is a write operation that creates new data in the database. Write · High update_rows This tool modifies existing data in a PostgreSQL database. Write · High pause_entity This tool modifies the status of advertising entities (keywords, audiences, creatives, ad groups) by pausing them.

4 of Postgres's 10 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Postgres

PolicyLayer is an MCP gateway — it sits between your AI agents and Postgres, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "delete_rows": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "insert_row": {
    "limits": [
      {
        "counter": "insert_row_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "audit_account_waste": {
    "limits": [
      {
        "counter": "audit_account_waste_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Postgres — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON POSTGRES →

Instant setup, no code required.

FULL CATALOGUE

All 10 Postgres tools

DESTRUCTIVE 1 tools

Destructive delete_rows DELETE rows matching a WHERE clause and return the deleted records

WRITE 3 tools

Write insert_row INSERT a single row into a table and return the inserted record Write pause_entity Pause a specific ad entity (keyword, audience, creative, or ad group). ⚠️ ACTION TOOL: this changes the statu Write update_rows UPDATE rows matching a WHERE clause and return affected records

READ 6 tools

Read audit_account_waste Audit the entire ad account for wasted spend. Scans all entities (keywords, audiences, creatives, ad groups) Read describe_table Show column names, data types, nullability and defaults for a table Read get_waste_detail Get a detailed daily breakdown for a single ad entity (keyword, audience, creative, ad group). Use after audi Read list_tables List all user tables in the public schema Read select_rows SELECT rows from a table with an optional WHERE clause and row limit Read simulate_reallocation Simulate the budget impact of pausing one or more entities, WITHOUT making any changes. Calculates how much d

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Postgres

Can an AI agent delete data through the Postgres MCP server? +

Yes. The Postgres server exposes 1 destructive tools including delete_rows. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Postgres? +

The Postgres server has 3 write tools including insert_row, pause_entity, update_rows. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Postgres.

How many tools does the Postgres MCP server expose? +

10 tools across 3 categories: Destructive, Read, Write. 6 are read-only. 4 can modify, create, or delete data.

How do I enforce a policy on Postgres? +

Register the Postgres MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Postgres tool call.

Deterministic rules across all 10 Postgres tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON POSTGRES →

Instant setup, no code required.

10 Postgres tools catalogued and risk-classified — across an index of 43,000+ MCP servers.