hedging8563/tokenlab-deepseek-harness-provider ↗★ 0
@tokenlabai/dsh-provider
TokenLab provider bundle for DeepSeek Harness with native Responses, Messages and Chat routing plus full multimodal and async tools.
AI 分析
将 TokenLab 平台的大模型能力集成至 DSH。适合使用 TokenLab 渠道、需要多模态支持和异步工具调用能力的开发者。
インストール
npx -p @deepseek-ai/dsh dsh plugin --profile web add github:hedging8563/tokenlab-deepseek-harness-providerドキュメント
README 全文を読む ↗Configuration
Optional environment variables:
| Variable | Default | Purpose |
|---|---|---|
TOKENLAB_API_KEY | none | Shared TokenLab credential for model routes, MCP tools, and async wait |
TOKENLAB_API_BASE | https://api.tokenlab.sh | MCP and async-task API root |
TOKENLAB_OPENAI_BASE_URL | https://api.tokenlab.sh/v1 | Responses and Chat adapter base URL |
TOKENLAB_ANTHROPIC_BASE_URL | https://api.tokenlab.sh | Messages adapter base URL; the adapter appends /v1/messages |
The bundle intentionally uses the MCP full profile with portable schemas for complete phase-one coverage. If context size matters more than full developer coverage, set TOKENLAB_MCP_TOOL_PROFILE=core or override the tokenlab-async-tools row in the profile's cordis.patch.yml.
Existing llm-pi-ai settings
DSH currently has one shared llm-pi-ai settings section, and a saved user section has higher precedence than bundle defaults. If you already configured providers on the Models page, that saved section can replace this bundle's three TokenLab routes. Merge the tokenlab-responses, tokenlab-messages, and tokenlab-chat blocks from this package's cordis.patch.yml into the saved llm-pi-ai.providers map. This is a current Harness configuration-ownership constraint, not a TokenLab routing fallback.