capsule AI-native Unix-like composition layer

capsule.yaml

1,538 bytes · 49 lines · capsule://quake0day/[email protected] raw on github

apiVersion: capsule.dev/v0.1
kind: Capsule
name: cverse-rag-inference-adapter
version: 0.1.0
type: adapter
purpose:
  summary: 'Provides a Retrieval Augmented Generation (RAG) engine adapter, allowing
    the core inference service to integrate knowledge retrieval capabilities for enhancing
    LLM responses.

    '
  owns:
  - RAG engine implementation
  - RAG service interface
  does_not_own:
  - The core inference service
  - Other AI model types (ASR, LLM, TTS, Avatar, VoiceLLM)
interfaces:
  provides:
  - kind: library
    name: rag-plugin
    description: RAG engine implementation conforming to the inference core's plugin
      interface.
dependencies:
  capsules:
  - name: cverse-inference-core
    version: '>=0.1.0'
agent:
  summary_for_ai: 'An AI agent working on this capsule would focus on improving the
    RAG engine''s retrieval accuracy, integrating new knowledge sources, or optimizing
    the RAG pipeline for performance. It requires knowledge of information retrieval,
    vector databases, and Python development.

    '
  avoid:
  - Modifying the core inference service or other AI model types.
verification:
  invariants:
  - The RAG engine must retrieve relevant information based on queries.
  - Plugins must adhere to the defined RAG plugin interface.
x-reuse:
  notes: 'The RAG engine might require configuration for specific vector databases,
    embedding models, or knowledge source connectors, which would need to be parameterized
    by the consumer.

    '
x-reconstruct:
  install: install.json