MCP Toolbox for Databases
BLOCKgrade F · trust 40/100MCP Toolbox for Databases is an open source MCP server for databases.
agentagentsaibigqueryclickhousecockroachdbdatabaseelasticsearchOverview
From the repository's own README, as read at the audited commit.
<div align="center"># MCP Toolbox for Databases<a href="https://trendshift.io/repositories/25495" target="_blank"><img src="https://trendshift.io/api/badge/repositories/25495" alt="googleapis%2Fmcp-toolbox | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>[](https://opensource.org/licenses/Apache-2.0)[](https://mcp-toolbox.dev/)[](https://discord.gg/Dmm69peqjh)[](https://medium.com/@mcp_toolbox)[](https://pypi.org/project/toolbox-core/)[](https://www.npmjs.com/package/@toolbox-sdk/core)[](https://pkg.go.dev/github.com/googleapis/mcp-toolbox-sdk-go)[](https://mvnrepository.com/artifact/com.google.cloud.mcp/mcp-toolbox-sdk-java)</div>MCP Toolbox for Databases is an open source Model Context Protocol (MCP) server that connects your AI agents, IDEs, and applications directly to your enterprise databases. <p align="center"><img src="docs/en/documentation/introduction/architecture.png" alt="architecture" width="50%"/></p>It serves a **dual purpose**:1. **Ready-to-use MCP Server (Build-Time):** Instantly connect Gemini CLI, Google Antigravity, Claude Code, Codex, or other MCP clients to your databases using our *prebuilt generic tool
Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add toolbox:1.12.0 -- docker run -i --rm us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.12.0:None
Exposed tools (2) 2 read · 0 write · 0 destructive
| Tool | Risk | Description |
|---|---|---|
basic | read | foo bar |
my-skill | read | flag fallback |
Details
- Source
- googleapis/genai-toolbox
- npm
@toolbox-sdk/server@1.12.0- PyPI
toolbox-server- Transports
- streamable-http
- Credentials it reads
ANTHROPIC_API_KEYDB_PASSGEMINI_API_KEYGOOGLE_API_KEYGOOGLE_GENAI_API_KEY- License
- Apache-2.0
- Stars
- 16,446 · pushed 0d ago
Trust audit
Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | FAIL |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (3 observation(s))
- Network
- declared (10 observation(s))
- Shell
- declared (5 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
dsn += "?secure=true&skip_verify=false"
- **Prompt Injection Defense**: Use pre-processing hooks to scan user inputs for known jailbreak patterns or malicious directives before they reach the model or tools.
back in. This catches prompt injection and jailbreak attempts.
injection / jailbreak detection](https://docs.cloud.google.com/model-armor/overview#ma-prompt-injection):
3. Under **Prompt injection and jailbreak detection**, enable the filter and set
# Prompt injection / jailbreak: blocked at ingress.
want: "postgres://u:[email protected]:5432/mydb",
creds = CredentialStrategy.manual_token(token="your-static-bearer-token")
token: "opaque-valid-string-exp",
token: "opaque-valid-array-aud",
token: "opaque-missing-active",
token: "header.payload.signature",
.gitmodules
.golangci.yaml
.browserslistrc
.lycheeignore
.gemini/styleguide.md
AGENTS.md
CLAUDE.md
skills/maintainer/reproduce-bug/references/DEVELOPER.md
skills/maintainer/review-prs/references/CONTRIBUTING.md
agent = importlib.import_module(module_path)
quickstart = importlib.import_module(module_path)
const envPath = path.resolve(__dirname, '../../../.env');
const localPath = path.resolve(__dirname, '../../../toolbox' + ext);
Gates applied: instruction_override, no_behavioural_pass.
Audited 2026-09-18 · audit v0.4.0 · source sha a7ab56d3105a · full audit: observations/trust-audit/mcp-server/googleapis__genai-toolbox.json · Report an issue or request a re-scan
Audit history
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-09-18 | a7ab56d3105a | BLOCK | F | 40 | first audit |
Alternatives
Other servers in the same categories, safer ones first.
More from googleapis
Questions
What is the MCP Toolbox for Databases MCP server?
MCP Toolbox for Databases is an open source MCP server for databases.
What tools does MCP Toolbox for Databases expose?
2 in total: 2 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is MCP Toolbox for Databases safe to connect to an agent?
No — not without reading the findings first. The audit graded it F (40/100) and found 6 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What credentials does MCP Toolbox for Databases need?
It reads ANTHROPIC_API_KEY, DB_PASS, GEMINI_API_KEY, GOOGLE_API_KEY and GOOGLE_GENAI_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does MCP Toolbox for Databases run?
It speaks streamable-http, so it runs as a service you connect to over the network. It is published on npm as @toolbox-sdk/server at 1.12.0.
How current is this page?
The grade is for one exact copy of the source (a7ab56d3105a), read on 2026-09-18. The repository is watched and re-audited when it changes.
Provenance: OBSERVED · read 2026-09-18 · job trust-audit-2026-09-18