Skip to content
MCP server

Vericto: database guardrails for AI agents

Check SQL written by AI agents against your Vericto database rules before it runs

Install
Pricing
Freemium
Platforms
API, CLI, self-hosted
Runs
Remotely
Version
1.1.0
Listing updated
2 days ago

About Vericto: database guardrails for AI agents

Vericto: database guardrails for AI agents is a freemium MCP server for databases. It runs as a hosted remote server, so there is nothing to install locally.

Vericto's MCP integration puts rule checking at the moment an agent writes SQL. A connected agent calls check_sql and gets one of four verdicts, BLOCKED, FLAGGED, MONITORED or ALLOWED, with the rule that fired and a suggested fix. The docs describe it as the earliest of three checkpoints, ahead of the CLI check before merge and the TCP proxy that blocks in production, all using the same deterministic AST engine.

There are three ways to connect. A hosted account server signs in with OAuth and acts with your role. A hosted API-key server is meant for agents and CI and should use a key scoped only to ci_dryrun:execute. A local vericto-db server runs in Docker against your own PostgreSQL and offers execute_sql, describe_schema, check_sql and whoami without handing the agent the database credential. Setup snippets cover Claude Code, Claude Desktop, Cursor, VS Code, Kiro and Antigravity.

MCP and CLI checks count against the plan: 1K a month on Free, 10K on Builder and unlimited on Team.

Install

Run in a terminal

claude mcp add --transport http mcp https://mcp.vericto.com/mcp

Add to Settings → Connectors → Add custom connector

https://mcp.vericto.com/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "mcp": {
      "url": "https://mcp.vericto.com/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "mcp": {
      "type": "http",
      "url": "https://mcp.vericto.com/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "mcp": {
      "serverUrl": "https://mcp.vericto.com/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "mcp": {
      "type": "http",
      "url": "https://mcp.vericto.com/mcp"
    }
  }
}

Tools this server provides (4)

Taken from the project documentation; the running server may differ.

check_sql
Checks SQL against workspace rules and returns a verdict with the rule and a suggested fix.
execute_sql
Runs SQL against the connected PostgreSQL database (local vericto-db server).
describe_schema
Describes the database schema (local vericto-db server).
whoami
Reports the identity the server is acting as (local vericto-db server).

What it does

Four-level verdicts
Each check returns BLOCKED, FLAGGED, MONITORED or ALLOWED with the rule and a fix.
Shared rule engine
Uses the same AST engine and workspace rules as the CLI and TCP proxy.
OAuth or API key
Sign in through the browser, or use a narrowly scoped key for CI.
Local Docker mode
vericto-proxy mcp reaches your PostgreSQL without exposing credentials to the agent.
Read-only role guidance
The docs walk through creating a read-only database role for the local server.

Good for

  • Validate migrations while a coding agent drafts them
  • Stop destructive queries before they reach a repository
  • Give an agent schema access to PostgreSQL without sharing the connection string
  • Apply one rule set to agents, CI and production traffic

Pricing

Freemium, from $79 a month

Free plan with 1K SQL checks/month; Builder $79/mo, Team $249/mo, Enterprise from $2,500/mo

Prices change. Check vericto.com before you buy.

Vericto: database guardrails for AI agents FAQ

What does a check return?

A verdict (BLOCKED, FLAGGED, MONITORED or ALLOWED), the rule that fired and a suggested fix.

Does the agent see my database password?

With the local vericto-db server, no. The URL is read from a file mounted into Docker and the agent never receives it.

How many checks are included?

Free includes 1K SQL checks a month for CLI and MCP, Builder 10K, and Team unlimited.

Which key scope should I use?

The docs recommend a key with only the ci_dryrun:execute scope.

Is Vericto: database guardrails for AI agents free?

Vericto: database guardrails for AI agents has a free plan, with paid plans for more usage or features. Paid plans start at $79 a month.

How do I add Vericto: database guardrails for AI agents to Claude Code?

Run this command in your terminal: claude mcp add --transport http mcp https://mcp.vericto.com/mcp

What are the best alternatives to Vericto: database guardrails for AI agents?

The closest alternatives to Vericto: database guardrails for AI agents are REA, Agent Roast, Midplane, Leash and Tinplate Edge AI Safety, Security & EU Compliance.

Alternatives to Vericto: database guardrails for AI agents

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Agent Roast

    MCP server

    Offline review that flags agent changes which weaken tests, assertions or CI

    Runs locally
    • Open source
    • 4 Hacker News points
    • Developer tools
  3. Midplane

    MCP server

    Postgres gateway for AI agents with table grants, column masking and approval for writes

    Runs locally
    • Open source
    • 5 GitHub stars
    • Databases
  4. Leash

    MCP server

    Review what your AI touched, approve risky actions and undo supported changes

    • 1 tool
    • Remote
    • Free
    • 1 GitHub star
    • Security
  5. SecureCloud

    MCP server

    SentinelOne and IT security catalog, prices, quotes and checkout links for Israel

    • 7 tools
    • Remote
    • Free
    • Security