https://github.com/gonkagate/qwen-code-setup
Install GonkaGate for Qwen Code with one npx command. Secure setup CLI for @qwen-code/qwen-code OpenAI-compatible provider config, live model discovery, and safe API key storage.
https://github.com/gonkagate/qwen-code-setup
ai-cli ai-coding ai-models api-key cli coding-agent custom-provider developer-tools gonkagate llm model-discovery npx openai-api openai-compatible qwen qwen-code qwen-code-cli secure-setup typescript
Last synced: 1 day ago
JSON representation
Install GonkaGate for Qwen Code with one npx command. Secure setup CLI for @qwen-code/qwen-code OpenAI-compatible provider config, live model discovery, and safe API key storage.
- Host: GitHub
- URL: https://github.com/gonkagate/qwen-code-setup
- Owner: GonkaGate
- License: apache-2.0
- Created: 2026-06-12T13:05:36.000Z (about 1 month ago)
- Default Branch: main
- Last Pushed: 2026-07-06T23:44:06.000Z (7 days ago)
- Last Synced: 2026-07-07T01:10:59.369Z (7 days ago)
- Topics: ai-cli, ai-coding, ai-models, api-key, cli, coding-agent, custom-provider, developer-tools, gonkagate, llm, model-discovery, npx, openai-api, openai-compatible, qwen, qwen-code, qwen-code-cli, secure-setup, typescript
- Language: TypeScript
- Homepage: https://gonkagate.com
- Size: 369 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- License: LICENSE
- Security: docs/security.md
- Agents: AGENTS.md
Awesome Lists containing this project
README
# Qwen Code Setup for GonkaGate
Configure Qwen Code to use GonkaGate as an OpenAI-compatible provider with one
developer-friendly `npx` command.
```bash
npx @gonkagate/qwen-code-setup
```
[](https://gonkagate.com/en?utm_source=github&utm_medium=referral&utm_campaign=qwen_code_setup&utm_content=readme_badge_website)
[](https://gonkagate.com/en/docs?utm_source=github&utm_medium=referral&utm_campaign=qwen_code_setup&utm_content=readme_badge_docs)
[](https://gonkagate.com/en/register?utm_source=github&utm_medium=referral&utm_campaign=qwen_code_setup&utm_content=readme_badge_api_key)
`@gonkagate/qwen-code-setup` is a public CLI installer for developers who want
Qwen Code to route chat completions through the GonkaGate API without manually
editing Qwen settings JSON.
## Quick Start
```bash
npx @gonkagate/qwen-code-setup
```
You need:
- Node.js `>=22.14.0`
- Qwen Code installed as `qwen`
- a GonkaGate API key from the dashboard
The runtime is implemented. It writes managed Qwen Code settings after
validating Qwen Code, collecting the key through safe inputs, checking the
GonkaGate model catalog, verifying the local result, and keeping output
redacted.
## What It Does
The happy path is:
1. Detects the local `qwen` binary and records its version.
2. Collects or reuses `GONKAGATE_API_KEY` without printing the secret.
3. Calls authenticated `GET https://api.gonkagate.com/v1/models`.
4. Uses that live model response for defaulting, selection, and validation.
5. Writes the GonkaGate provider into `modelProviders.openai[]`.
6. Sets `security.auth.selectedType = "openai"` and `model.name`.
7. Stores the durable key reference at `settings.env.GONKAGATE_API_KEY`.
8. Creates backups, rolls back on failure, and reports current-session
environment shadowing when relevant.
Use `--dry-run` to inspect planned writes before changing local Qwen Code
settings. Use `--json` for machine-readable, redacted output.
## Supported Models
The installer has no checked-in user-facing model catalog. After collecting the
GonkaGate API key, it calls authenticated
`GET https://api.gonkagate.com/v1/models` and uses the returned model ids and
names as the source of truth for the picker, `--model` validation, default
selection, and Qwen Code provider writes.
## Known Qwen Code Baseline
The compatibility baseline is `@qwen-code/qwen-code` `0.18.0`, audited
on June 12, 2026.
Newer Qwen Code versions are not blocked solely because they are newer; the
setup and verification checks surface actual incompatibilities.
The current Qwen Code integration points are:
- binary: `qwen`
- package: `@qwen-code/qwen-code`
- package engine: `>=22.0.0`
- user settings: `~/.qwen/settings.json`
- `QWEN_HOME` user settings: `/settings.json`
- provider family: `modelProviders.openai[]`
- auth selection: `security.auth.selectedType = "openai"`
- active model selection: `model.name`
- provider key lookup: `envKey`, resolved from Qwen Code environment loading
- model discovery: authenticated `GET https://api.gonkagate.com/v1/models`
after API-key collection
The audited `0.18.0` source marks `modelProviders` with `replace` merge
semantics. User-scope provider catalog writes are managed in user settings.
Project scope writes only activation settings and blocks if project
`modelProviders` would hide user-managed providers.
## Security
The installer is designed for local developer machines and keeps secrets out of
plain command history.
Allowed secret inputs:
- hidden interactive prompt
- `GONKAGATE_API_KEY`
- `--api-key-stdin`
Not allowed:
- plain `--api-key`
- shell profile mutation
- repository-local secret storage
The v1 durable secret target is user-level
`settings.env.GONKAGATE_API_KEY` inside the active Qwen settings file. Unlike
OpenCode, the audited Qwen Code docs describe `envKey` and environment loading
rather than an OpenCode-style `{file:...}` provider secret binding.
## Verification And Safety
Default setup success is based on bounded local verification, not a live model
call. `--verify-live` is available only as an explicit opt-in and may spend
quota or depend on provider/network availability.
## Local Development
Install and run the full contract suite:
```bash
npm ci
npm run ci
```
Useful focused checks:
```bash
npm run build
npm run typecheck
npm run test
npm run format:check
npm run package:check
npm pack --dry-run
```
Run the CLI from source:
```bash
npm run dev -- --dry-run
```
## Release Flow
This repository uses Release Please and npm trusted publishing.
For changes that should create a release, use a releasable Conventional Commit:
```bash
feat: improve qwen code developer onboarding
```
After that commit lands on `main`, Release Please opens a release PR that bumps
`package.json`, `.release-please-manifest.json`, `CHANGELOG.md`, and
`src/constants/contract.ts`. When that release PR is merged, GitHub Actions
publishes to npm with provenance.
Use `fix:` for patch releases and `feat:` for minor releases. Non-releasable
commits such as `docs:` are useful for internal cleanup, but they should not be
used when the change is intentionally testing the release pipeline.
## Docs
- [Product spec](docs/specs/qwen-code-setup-prd/spec.md)
- [How it works](docs/how-it-works.md)
- [Security](docs/security.md)
- [Model validation](docs/model-validation.md)
- [Architecture decisions](docs/architecture-decisions.md)
- [Troubleshooting](docs/troubleshooting.md)