loongport

loongport configures one LoongPort OpenAI-compatible route for DeepSeek Harness (DSH).

Install

Run it directly with npm:

LOONGPORT_API_KEY='sk-…' npx loongport dsh setup \
  --base-url https://relay.example.com/v1 \
  --model model-id \
  --write

The API key is read only from LOONGPORT_API_KEY; do not place it on the command line. You can repeat --model to register more than one model.

By default, the command performs a dry run. Add --write to update DSH's configuration. It uses $DSH_HOME when set, otherwise ~/.dsh, and writes:

• settings.yaml: one selected llm-pi-ai.providers.<route> entry
• .credentials.yaml: the selected credential name and API key (owner-only)

Other DSH settings, provider routes, and credentials are preserved. The command uses DSH's built-in llm-pi-ai transport; it does not implement its own streaming, retries, or tool calls.

Options

--base-url <url>              OpenAI-compatible HTTP(S) endpoint (required)
--model <id>                  Model to expose; repeat for multiple models (required)
--route <name>                DSH provider route (default: loongport)
--credential-name <name>      DSH credential key (default: LOONGPORT_API_KEY)
--write                       Apply changes; omitted means dry run

The initial release supports manually provisioned OpenAI-compatible endpoints. It does not automate third-party browser login, relay discovery, desktop-app bridging, or automatic tier switching.

Maintainers

Agents and maintainers should read the agent guide before changing the package and follow the release runbook before publishing.

License

MIT