Add Claude Code support (additive; Codex path unchanged)#5
Open
WenyuChiou wants to merge 1 commit into
Open
Conversation
Make the render step tool-agnostic so the skill runs on Claude Code too, without changing the 小黑 identity or Codex behaviour. - SKILL.md section 3: render via whatever image tool exists (Codex image_gen, an image-gen MCP, or emit ready-to-paste prompts). Widen the description to also trigger on English requests. - references/prompt-template.md: note that image-edit prompts need an edit/inpaint-capable tool. - README.md: add a "在 Claude Code 使用" install section. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Hi Ian — thanks for this skill, the 小黑 visual language is lovely.
This is a small additive PR so the skill also runs on Claude Code, with no change to the 小黑 identity, the Chinese-first focus, or the Codex behaviour.
What it changes
ian-xiaohei-illustrations/SKILL.md— the render step (§3) was hard-coded to Codex's built-inimage_gen. It's now tool-agnostic: useimage_genif present (Codex), else an image-generation MCP (e.g. Claude'smcp__image-gen__generate_image, 16:9 / png), else emit ready-to-paste prompts. Codex behaves exactly as before. Also widened thedescriptionso it triggers on English requests too.ian-xiaohei-illustrations/references/prompt-template.md— a note that the image-edit prompts need an edit/inpaint-capable tool (Codeximage_gen); pure text-to-image MCPs should regenerate-with-fewer-labels instead.README.md— a short "在 Claude Code 使用" section (install + the tool-agnostic note).Why
The skill's layout is byte-identical to Anthropic's Agent Skills format, so it already loads in Claude Code — only the render call was platform-specific. This makes it run end-to-end on both Codex and Claude.
It also answers the two open issues:
Kept deliberately minimal and identity-neutral. Full disclosure: I also maintain a more divergent fork (a fully-English variant + bring-your-own-mascot), but that is not part of this PR — this is just "your skill now runs on Claude too." Happy to adjust anything.
🤖 Generated with Claude Code