MCP Toolbox SDKs for JSSAFE
Javascript SDK for interacting with the MCP Toolbox for Databases.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://opensource.org/licenses/Apache-2.0) [](https://mcp-toolbox.dev/) [](https://discord.gg/Dmm69peqjh) [](https://medium.com/@mcp_toolbox)
This repository contains JavaScript SDKs for MCP Toolbox. These SDKs allow you to load and use tools defined in your MCP Toolbox server as standard JS functions or objects within your Agentic applications.
For comprehensive guides, tutorials, and advanced configuration, visit the Main Documentation Site.
- Available Packages
- Quick Start
- Contributing
- License
- Support
Available Packages
This repository hosts the following TS packages. Visit the Main Documentation Site for detailed installation and usage instructions:
Quick Start
- Set up the Toolbox Service: Ensure you have a running MCP Toolbox server. Follow the [Toolbox Server Getting Started Gu
ed3ab3bd881dOBSERVED · 2026-10-12Connect
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.
claude mcp add core -- npx -y @toolbox-sdk/[email protected]
{
"mcpServers": {
"core": {
"command": "npx",
"args": [
"-y",
"@toolbox-sdk/[email protected]"
]
}
}
}Exposed tools (52)
50 read · 2 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
age | read | User age |
authTool | read | Auth required |
authTool2026 | read | Auth required 2026 |
authToolLegacyOn2026 | read | Legacy auth params on 2026 version |
count | read | count number |
create_user | write | Create a new user |
faultyParam | read | This param has an unhandled type |
get_time | read | Get the current time |
get_weather | read | Get the weather for a location |
id | read | An identifier |
innerArray | read | Inner array of integers |
input | read | input string |
intItem | read | integer item |
int_item | read | item_desc |
isActive | read | User status |
item_name | read | item_desc |
limit | read | query limit |
log_event | read | Log an event |
matrix | read | A matrix of numbers |
metadata | read | Untyped metadata object |
optionalInt | read | An optional integer |
optionalObject | read | An optional object |
optionalParam | read | optional |
optionalString | read | An optional string |
otherTool | read | desc2 |
outerArray | read | Outer array |
param1 | read | String param |
process_data | read | Process some data |
requiredInt | read | A required integer |
requiredParam | read | required |
requiredString | read | A required string |
sample_tool | read | Sample |
scores | read | Map of names to scores |
secureTool | read | A secure tool |
strictObject | read | Object with no additional properties allowed |
tags | read | List of tags |
targetTool | read | desc |
testArray | read | An array of strings |
testArrayInt | read | An array of integers |
testBool | read | A boolean |
testFloat | read | A float |
testInt | read | An integer |
testNumber | read | A number |
testParam | read | A param |
testString | read | A string |
testTool | read | Test Description |
test_tool | read | A mock tool for testing |
typedMap | read | A map of strings to integers |
unsupported_test | read | Test unsupported types |
untypedMap | read | An untyped map |
update_settings | write | Update user settings |
username | read | User login name |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (4 observation(s))
- Network
- declared (2 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (13)
.eslintignore
.prettierrc.cjs
.release-please-manifest.json
.prettierrc.cjs
} from '../../src/toolbox_adk/index.js';
import {ZodManifest, Protocol} from '../../protocol.js';import {logApiError, ProtocolNegotiationError} from '../../errorUtils.js';import {warnIfHttpAndHeaders} from '../../utils.js';import {VERSION} from '../../version.js';autoprefixer, postcss, postcss-cli
@google-cloud/secret-manager, @google-cloud/storage, @jest/globals, @types/fs-extra, @types/jest, @types/node, @types/tmp, cross-env
@google/adk, @google/genai, @toolbox-sdk/core, axios, openapi-types, rimraf, zod
axios, google-auth-library, rimraf, zod
Gates applied: no_behavioural_pass.
ed3ab3bd881dfull audit observations/trust-audit/mcp-server/googleapis__mcp-toolbox-sdk-js.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-12 | ed3ab3bd881d | SAFE | B | 89 | first audit |
Questions
What is the @toolbox-sdk/core MCP server?
Javascript SDK for interacting with the MCP Toolbox for Databases.
What tools does @toolbox-sdk/core expose?
52 in total: 50 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is @toolbox-sdk/core safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does @toolbox-sdk/core need?
No credential environment variables were found in its source, so it appears to need none.
How current is this page?
The grade is for one exact copy of the source (ed3ab3bd881d), read on 2026-10-12. The repository is watched and re-audited when it changes.