memex/notes/org-skill-provider-openai.org

SKILL: OpenAI Provider Agent (Universal Literate Note)

• Overview
• Phase A: Demand (PRD)
  • 1. Purpose
  • 2. User Needs
  • 3. Success Criteria
    • API Authentication via Bearer Token
    • Chat Payload Construction
    • Choice Extraction Verification
• Phase B: Blueprint (PROTOCOL)
  • 1. Architectural Intent
  • 2. Semantic Interfaces
• Phase D: Build (Implementation)
  • Request Execution
  • Model Discovery
• Registration

Overview

The OpenAI Provider Agent integrates OpenAI's GPT models as a pluggable System 1 (neural) backend. It provides industry-standard processing capabilities for reasoning and content generation.

Phase A: Demand (PRD)

1. Purpose

Define the interface for reliable communication with the OpenAI Chat Completions API.

2. User Needs

• Compatibility: Full implementation of the Chat Completions protocol.
• Reliability: Secure management of `$OPENAI_API_KEY`.
• Optimized Inference: Configurable temperature and model selection (GPT-4o, etc.).
• Resilience: Graceful handling of timeouts and errors.

3. Success Criteria

TODO API Authentication via Bearer Token

TODO Chat Payload Construction

TODO Choice Extraction Verification

Phase B: Blueprint (PROTOCOL)

1. Architectural Intent

Interfaces for executing neural completion requests via OpenAI's Chat API.

2. Semantic Interfaces

(defun execute-openai-request (prompt system-prompt)
  "Executes a completion request via the OpenAI API.")

(defun get-openai-models ()
  "Returns supported GPT models and their context limits.")

Phase D: Build (Implementation)

Request Execution

(defun execute-openai-request (prompt system-prompt)
  (let ((api-key (uiop:getenv "OPENAI_API_KEY")))
    (unless api-key (return-from execute-openai-request "ERROR: Key missing"))
    (let ((model (get-config-attribute :LLM_MODEL_OPENAI "gpt-4o")))
      ;; Physical API call logic (mocked for refactor)
      (format nil "Executing OpenAI request on ~a" model))))

Model Discovery

(defun get-openai-models ()
  '((:id "gpt-4o" :context "128k")
    (:id "gpt-4-turbo-preview" :context "128k")
    (:id "gpt-3.5-turbo" :context "16k")))

Registration

(defskill :skill-provider-openai
  :priority 100
  :trigger (lambda (context) nil)
  :neuro (lambda (context) nil)
  :symbolic (lambda (action context) action))