TenetLLMAdapters

Cloud LLM adapter package for the Tenet platform.

This repository provides four adapters that register through the tenet.llm_adapters entry-point group and implement both generation and live model discovery:

• anthropic -> AnthropicAdapter
• openai-compatible -> OpenAIAdapter
• google -> GoogleAdapter
• cohere -> CohereAdapter

Platform Context

• Requirements: SRS_TENETLLMADAPTERS
• Architecture: ARCH_TENETLLMADAPTERS
• Realization: PLAN_TENETLLMADAPTERS
• Verification: VER_TENET_LLM_ADAPTERS
• Governance: TenetOS

Install

Base install includes no cloud SDK dependencies:

pip install tenet-llm-adapters

Install provider-specific extras as needed:

pip install tenet-llm-adapters[anthropic]
pip install tenet-llm-adapters[openai]
pip install tenet-llm-adapters[google,cohere]
pip install tenet-llm-adapters[all]

TenetCore usage

Set the provider in a TenetCore backend config and point API keys to environment variables:

llm:
  backends:
    claude:
      provider: anthropic
      api_key: "${ANTHROPIC_API_KEY}"
    shared-openai:
      provider: openai-compatible
      api_key: "${OPENAI_API_KEY}"
    gemini:
      provider: google
      api_key: "${GOOGLE_API_KEY}"
    cohere:
      provider: cohere
      api_key: "${COHERE_API_KEY}"

Entry points

[project.entry-points."tenet.llm_adapters"]
anthropic = "tenet_llm_adapters._anthropic:AnthropicAdapter"
openai-compatible = "tenet_llm_adapters._openai:OpenAIAdapter"
google = "tenet_llm_adapters._google:GoogleAdapter"
cohere = "tenet_llm_adapters._cohere:CohereAdapter"