Skip to content

feat(agentized): add Gemini image provider + QC auto-fix skill + KIE.…#314

Open
sabalioglu wants to merge 1 commit into
calesthio:mainfrom
sabalioglu:main
Open

feat(agentized): add Gemini image provider + QC auto-fix skill + KIE.…#314
sabalioglu wants to merge 1 commit into
calesthio:mainfrom
sabalioglu:main

Conversation

@sabalioglu

Copy link
Copy Markdown

…AI tools

  • tools/graphics/gemini_image.py — Gemini 2.5 Flash Image (default) + 3 Pro Image Preview, ported from VRSEN/OpenSwarm pattern. Distinct from google_imagen (which covers Imagen 4.0 family).
  • skills/creative/image-qc-autofix.md — 5-bullet QC pass + max-1-fix loop pattern. Complements EP gating for single-shot images outside full pipelines.
  • skills/creative/image-provider-usage.md — registry update: gemini_image listed, scene-type table updated for text-heavy / multi-constraint / iterative use cases.
  • lib/kie_client.py — shared KIE.AI HTTP client (Pattern A unified market API).
  • tools/graphics/kie_{gpt_image,nano_banana}.py + tools/video/kie_{kling,seedance}.py — KIE.AI provider tools.
  • tools/audio/qwen3_tts.py — Qwen3 TTS provider.
  • WORKSPACE.md — workspace conventions doc.
  • .gitignore — exclude .env.backup-* and .claude-flow/.

Summary

Related issue

Closes #

Changes

Testing

Checklist

  • The change is focused on a single logical concern.
  • I ran the relevant tests locally (make test-contracts / make test) where applicable.
  • I updated docs/README if behavior or usage changed.
  • No unrelated files (build artifacts, local config) are included in the diff.

…AI tools

- tools/graphics/gemini_image.py — Gemini 2.5 Flash Image (default) + 3 Pro Image Preview, ported from VRSEN/OpenSwarm pattern. Distinct from google_imagen (which covers Imagen 4.0 family).
- skills/creative/image-qc-autofix.md — 5-bullet QC pass + max-1-fix loop pattern. Complements EP gating for single-shot images outside full pipelines.
- skills/creative/image-provider-usage.md — registry update: gemini_image listed, scene-type table updated for text-heavy / multi-constraint / iterative use cases.
- lib/kie_client.py — shared KIE.AI HTTP client (Pattern A unified market API).
- tools/graphics/kie_{gpt_image,nano_banana}.py + tools/video/kie_{kling,seedance}.py — KIE.AI provider tools.
- tools/audio/qwen3_tts.py — Qwen3 TTS provider.
- WORKSPACE.md — workspace conventions doc.
- .gitignore — exclude .env.backup-* and .claude-flow/.
@sabalioglu sabalioglu requested a review from calesthio as a code owner July 6, 2026 19:27

@calesthio calesthio left a comment

Copy link
Copy Markdown
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you for the contribution. There are useful ideas here, but I need to request changes before this can be reviewed safely for merge.

Current blockers:

  • The PR is in a dirty/conflicting merge state.
  • The scope is too broad for one PR: Gemini image provider, a shared KIE client, four KIE image/video providers, Qwen3 TTS, two creative skills, workspace docs, and .gitignore changes. The review guide calls out this kind of mixed provider/runtime/docs bundle as unsafe to approve because each provider needs its own contract review.
  • The PR template is effectively incomplete: testing/checklist sections are blank/unchecked.
  • There are no focused mocked tests for the new paid/network/runtime tools. Provider additions should cover metadata/contract, registry discovery, status behavior, API error handling, and mocked happy paths where possible.
  • User setup docs are incomplete for the new surfaces. For example, the tools read KIE_AI_API_KEY / KIEAI_API_KEY, and qwen3_tts has a local system-Python/runtime dependency, but .env.example / provider docs are not updated at the same level as the behavior.

Verification I ran:

python -m py_compile lib/kie_client.py tools/graphics/gemini_image.py tools/graphics/kie_gpt_image.py tools/graphics/kie_nano_banana.py tools/video/kie_kling.py tools/video/kie_seedance.py tools/audio/qwen3_tts.py
# passed

# Registry discovery check:
# image_generation: gemini_image, kie_gpt_image, kie_nano_banana
# video_generation: kie_kling, kie_seedance
# tts: qwen3_tts

git diff --check origin/main...HEAD
# clean

So the code compiles and discovers, but that is not enough for six new provider/runtime tools. Please split this into focused PRs, starting with one provider family at a time, resolve the merge conflicts, complete the PR description, and add mocked provider tests plus setup docs for each new tool. I did not find an actionable security issue in the reviewed diff, but the scope/test/doc gaps are merge blockers.

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.

2 participants