Skip to content

Fix image provider selection flow - #1

Open
laiyangjun-dotcom wants to merge 2 commits into
AI-flower:mainfrom
laiyangjun-dotcom:codex/fixed-image-provider-selection
Open

Fix image provider selection flow#1
laiyangjun-dotcom wants to merge 2 commits into
AI-flower:mainfrom
laiyangjun-dotcom:codex/fixed-image-provider-selection

Conversation

@laiyangjun-dotcom

Copy link
Copy Markdown

Summary

  • Require users to choose a fixed image provider (gemini-native or openai-images) during setup instead of auto-switching at runtime.
  • Add provider-isolated key/model resolution: Gemini uses GEMINI_API_KEY; OpenAI-compatible uses IMAGE_GEN_API_KEY / OPENAI_API_KEY.
  • Update image generation, preflight, install flow, runtime template, and skill docs to reflect the fixed-provider workflow.

Validation

  • python3 -m py_compile scripts/image.py scripts/preflight.py
  • bash -n install.sh
  • Provider isolation smoke test via direct scripts/image.py import
  • IMAGE_GEN_PROTOCOL=auto IMAGE_GEN_API_KEY=dummy python3 scripts/image.py --check exits 2
  • IMAGE_GEN_PROTOCOL=gemini-native GEMINI_API_KEY=dummy python3 scripts/image.py --check exits 0
  • IMAGE_GEN_PROTOCOL=openai-images IMAGE_GEN_API_KEY=dummy python3 scripts/image.py --check exits 0

Notes

  • Runtime secrets are intentionally excluded. config/runtime.env is not included; only config/runtime.env.example is updated.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant