ChangesetsSAFE
🧡 Turn your API spec into type-safe code. Generate TypeScript types, API clients, hooks, validators and mocks from a spec such as OpenAPI.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Hello and welcome! This folder has been automatically generated by @changesets/cli, a build tool that works with multi-package repos, or single-package repos to help you version and publish your code. You can find the full documentation for it in our repository
We have a quick list of common questions to get you started engaging with this project in our documentation
fda6963caa32OBSERVED · 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. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add claude-plugin --env KUBB_AGENT_TOKEN=${KUBB_AGENT_TOKEN} --env KUBB_TOKEN=${KUBB_TOKEN} -- npx -y @kubb/[email protected]{
"mcpServers": {
"claude-plugin": {
"command": "npx",
"args": [
"-y",
"@kubb/[email protected]"
],
"env": {
"KUBB_AGENT_TOKEN": "${KUBB_AGENT_TOKEN}",
"KUBB_TOKEN": "${KUBB_TOKEN}"
}
}
}
}Exposed tools (13)
8 read · 5 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
connect | read | Connect this project to Kubb Studio and generate from the browser. |
foo | read | |
generate | read | Generate TypeScript types, API clients, React Query hooks, Zod schemas, and more from an OpenAPI specification. Reads kubb.config.ts by default. Pass an OpenAPI file path as the first argument to override the input without editing the config. |
init | write | Scaffold a kubb.config.ts and install plugins for code generation from an OpenAPI spec. Run without flags for interactive setup, or pass --input, --output, and --plugins to skip the prompts. |
login | read | Pair this machine with Kubb Studio without connecting. |
logout | read | Forget the stored Kubb Studio token. |
mcp | write | Start a Model Context Protocol (MCP) server that exposes Kubb code generation as tools for AI assistants. Once running, configure your AI client (Claude, Cursor, Windsurf, etc.) to connect to it — the assistant can then call kubb generate directly without leaving the chat. |
snapshot | write | Generate and publish a Kubb Studio snapshot from CI. |
start | write | Keep this project connected to Kubb Studio in the background. |
status | read | Show the project worker, Kubb Studio pairing, and saved permissions. |
stop | write | Stop this project |
studio | read | Connect this project to Kubb Studio, or manage its pairing and snapshots. |
validate | read | Parse and validate an OpenAPI/Swagger file for structural correctness. Reports schema errors, missing required fields, and malformed references. Use this before running generate to catch spec issues early. |
Trust audit
SAFEgrade B · trust 88/100 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 | WARN |
| 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 (3 observation(s))
- Network
- declared (4 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (19)
.claude/skills
.all-contributorsrc
README.md
const accept = createHash('sha1').update(`${req.headers['sec-websocket-key']}258EAFA5-E914-47DA-95CA-C5AB0DC85B11`).digest('base64')export const hashOf = (content: string) => createHash('sha1').update(content).digest('hex').slice(0, 16)import { adapterOas } from '../../adapter.ts'import type { SchemaObject } from '../../types.ts'import type { SchemaObject } from '../../types.ts'import { isDiscriminator, isReference } from '../../oas.ts'import type { Refs } from '../../refs.ts'studioUrl: `http://127.0.0.1:${port}`,['a URL', 'http://127.0.0.1:8080/internal'],
[{ studioUrl: 'http://127.0.0.1:3000' }, 'http://127.0.0.1:3000/api/agent/token'],it.each(['http://studio.example.com', 'http://10.0.0.1', 'http://[fe80::1]'])('refuses plaintext to %s', (studioUrl) => {jiti
@changesets/changelog-github, @changesets/cli, bun-types, taze, turbo
@readme/openapi-parser, @scalar/openapi-upgrader, @types/json-schema, openapi-types
@clack/prompts, @gunshi/plugin-dryrun, chokidar, gunshi, jiti, unconfig, verkit
unstorage, vue
Gates applied: no_behavioural_pass.
fda6963caa32full audit observations/trust-audit/mcp-server/kubb-labs__kubb.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 | fda6963caa32 | SAFE | B | 88 | first audit |
Questions
What is the @kubb/mcp MCP server?
🧡 Turn your API spec into type-safe code. Generate TypeScript types, API clients, hooks, validators and mocks from a spec such as OpenAPI.
What tools does @kubb/mcp expose?
13 in total: 8 read-only, 5 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is @kubb/mcp safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (88/100). 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 @kubb/mcp need?
It reads KUBB_AGENT_TOKEN and KUBB_TOKEN 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 current is this page?
The grade is for one exact copy of the source (fda6963caa32), read on 2026-10-12. The repository is watched and re-audited when it changes.