# AgentSQL

**Canonical:** https://apis.io/providers/agentsql-cn/  
**Website:** https://agentsql.cn/  
**APIs profiled:** 1

AgentSQL（中文名：智盾）是面向 AI Agent 的数据库安全网关与 MCP Server。模型发出的 PostgreSQL / MySQL 请求在到达数据库前，经过身份认证、SQL 解析、静态与动态风险检查、四态决策与受控执行；结果按配置脱敏，并在应用层留下持久审计。AgentSQL 自有问数、图表、仪表板和嵌入应用页面，模型生成的 SQL 继续经过数据库安全网关。已有 SQLBot 部署也可接入 AgentSQL 做数据库访问控制；完整社区版功能、全部页面和现有系统零改动替换仍待逐项验收。

## Kin Score — 12.8 / 100 (emerging)

Scored 2026-11-06 under rubric 0.27.0. Trend: flat (-3.2 from 16.0).

| Facet | Score |
|---|---|
| Discoverability | 57.1 |
| Contract Quality | 0.0 |
| Contract Governance | 0.0 |
| Operational Transparency | 7.9 |
| Developer Ergonomics | 22.6 |
| Access Clarity | 0.0 |

Regulatory layer — **Horizontal (data, software, accessibility, platform)**: 25.5 (matched via fallback).

## Agent readiness — 2.5 (human-only)

| Dimension | Value |
|---|---|
| Spec Presence | no |
| Agentic Access | no |
| Reversibility Documented | no |
| MCP Server | no |
| Auth Clarity | bearer |
| Idempotency | no |
| Error Semantics | no |
| OpenAPI Examples | no |
| Rate Limit Signal | no |
| Event Surface Described | no |
| Agent Skills | no |
| Well Known Catalog | no |
| Consent Identity | no |
| Agent Card | no |
| Dry Run Mode | no |
| Delegated Identity | no |
| Protected Resource Metadata | no |
| Dynamic Client Registration | no |
| Agentic Commerce | no |
| Pagination Ratio | na |
| Operation Distinctiveness | na |
| MCP Tool Quality | na |

## Access

Unknown — onboarding: unknown, pricing: unknown, trial: no (confidence: low).

## APIs (1)

- **AgentSQL API** — AgentSQL provides a database security gateway and autonomous query interface.

## Security (3)

- **Agentsql Cn Authentication** — 1 scheme
- **Agentsql Cn Domain Security** — TLSv1.3 · HSTS
- **Agentsql Cn Vulnerability Disclosure** — security.txt · contact published

## Tags

Database, Security, Artificial Intelligence, Open Source, Gateways

---

Profiled by [API Evangelist](https://apievangelist.com) and published on [APIs.io](https://apis.io/providers/agentsql-cn/). Scores are computed from the provider's own public artifacts under a published rubric.