Codewhale

Skip it — one-line setup

The script installs Codewhale from its official source, writes the gateway settings into the tool's own config file (backing up any existing one) and runs a live check. It asks for your gateway key, or take it from --key / --key-file. --uninstall restores exactly what it wrote.

curl -fsSL https://quantoport.com/agents/codewhale/setup | bash

1 · Install

Install Codewhale from its own official source.

curl -fsSL https://codewhale.net/install.sh | sh   # macOS/Linux; Windows: winget install HunterBown.CodeWhale; or npm install -g codewhale

2 · Configure — point it at QuantoPort

Gateway base URL https://quantoport.com; get your key from the console.

1) Create ~/.codewhale/config.toml (or point CODEWHALE_CONFIG_PATH / --config at one).
2) Select the built-in OpenAI-compatible provider and point it at our gateway:

provider = "openai"
default_text_model = "<model-id>"

[providers.openai]
api_key = "YOUR_API_KEY"
base_url = "https://quantoport.com/v1"

On the China endpoint use base_url https://api.shinlinkai.com/v1 (OpenAI wire only).
3) You can also store the key first: codewhale auth set --provider openai --api-key "YOUR_API_KEY", then edit [providers.openai].base_url.
4) Restart the model client to apply. Pick the model with /model; the TUI's /provider (or F3) also accepts a pasted Base URL + key.

3 · Verify

One request proves you are on our channel.

codewhale exec "reply with OK"

4 · Common pitfalls

• Put the endpoint under the [providers.openai] table, not the legacy top-level base_url, or the OpenAI-compatible route will not pick it up.
• If the gateway accepts /chat/completions but rejects /v1/chat/completions, add path_suffix = "/chat/completions" under [providers.openai].
• Editing the base URL in config.toml requires restarting the model client to take effect.
• A non-localhost http:// gateway needs DEEPSEEK_ALLOW_INSECURE_HTTP=1 at launch; https is unaffected.