[Feature] Add OPENAI Support for Github Copilot API #3992
Description
Problem We Want to Solve
As a user of Skyvern, I want to be able to use GitHub Copilot's API when ENABLE_OPENAI_COMPATIBLE is True, so that I can perform workflow tasks.
Describe the Solution You'd Like
.env
WHEN LLM_KEY is OPENAI_COMPATIBLE
AND SECONDARY_LLM_KEY is OPENAI_COMPATIBLE
AND ENABLE_OPENAI_COMPATIBLE is True
AND OPENAI_COMPATIBLE_MODEL_NAME is a valid model name from api.githubcopilot.com/models
AND OPENAI_COMPATIBLE_API_KEY is a valid API key
AND OPENAI_COMPATIBLE_API_BASE is api.githubcopilot.com
AND OPENAI_COMPATIBLE_SUPPORTS_VISION is True or False (depends on the model you're using I believe)
THEN I start my Skyvern application
THEN I go to create, update, or run a workflow
THEN I can run said workflow with the OPENAI_COMPATIBLE GitHub Copilot API
Alternatives (optional)
¯\_(ツ)_/¯
Additional context
Aider guide and gettign GitHub Copilot API token
Example .env
ENV='local'
ENABLE_OPENAI='false'
OPENAI_API_KEY=''
ENABLE_ANTHROPIC='false'
ANTHROPIC_API_KEY=''
ENABLE_AZURE='false'
AZURE_DEPLOYMENT=''
AZURE_API_KEY=''
AZURE_API_BASE=''
AZURE_API_VERSION=''
ENABLE_AZURE_GPT4O_MINI='false'
AZURE_GPT4O_MINI_DEPLOYMENT=''
AZURE_GPT4O_MINI_API_KEY=''
AZURE_GPT4O_MINI_API_BASE=''
AZURE_GPT4O_MINI_API_VERSION=''
ENABLE_GEMINI='false'
GEMINI_API_KEY=''
ENABLE_NOVITA='false'
NOVITA_API_KEY=''
LLM_KEY='OPENAI_COMPATIBLE'
SECONDARY_LLM_KEY='OPENAI_COMPATIBLE'
BROWSER_TYPE='chromium-headless'
MAX_SCRAPING_RETRIES='3'
MAX_NUM_SCREENSHOTS='3'
VIDEO_PATH='./videos'
BROWSER_ACTION_TIMEOUT_MS='5000'
MAX_STEPS_PER_RUN='100'
LOG_LEVEL='DEBUG'
LITELLM_LOG='CRITICAL'
DATABASE_STRING='postgresql+psycopg://skyvern@localhost/skyvern'
PORT='8000'
ANALYTICS_ID=''
ENABLE_LOG_ARTIFACTS='false'
ENABLE_VOLCENGINE='false'
ENABLE_OPENAI_COMPATIBLE='true'
OPENAI_COMPATIBLE_MODEL_NAME='claude-haiku-4.5'
OPENAI_COMPATIBLE_API_KEY='Nuh Uh'
OPENAI_COMPATIBLE_API_BASE='https://api.githubcopilot.com'
OPENAI_COMPATIBLE_SUPPORTS_VISION='true'
OPENAI_COMPATIBLE_MAX_TOKENS='16384'
LLM_CONFIG_MAX_TOKENS='8192'