StarAPI

OpenHarness

ANTHROPIC_BASE_URL / ANTHROPIC_API_KEY / ANTHROPIC_MODEL

OpenHarness speaks the Anthropic Messages protocol, so three environment variables are enough — or choose the Anthropic-compatible provider profile inside oh setup. Set ANTHROPIC_MODEL explicitly to an id from the catalogue: dated Anthropic names such as claude-haiku-4-5-20251001 are not served here. Give it a generous output budget, since thinking-capable models spend part of max_tokens before any text appears.

Code

export ANTHROPIC_BASE_URL="https://starseaapi.com"
export ANTHROPIC_API_KEY="sk-your-key"
export ANTHROPIC_MODEL="deepseek-v4.1-flash"
oh                     # or: oh -p "..." --output-format json

Quickstart

1. Create an API key — Sign up, then create a token in the console. A key works across the whole catalogue; scope and quota can be restricted per key.
2. Point your client at the endpoint — Set base_url to https://starseaapi.com/v1 and use the key. Any OpenAI-compatible SDK, IDE or desktop client works.
3. Pick a model — Start with a low-cost tier for prototyping. Every id in the catalogue can be swapped without touching code.
4. Watch cost, not just quality — Output tokens are usually priced at a multiple of input, and reasoning models emit thinking tokens. Keep stable prompt prefixes first so caching applies.

Integrations