Skip to content

Add Higgsfield skills#404

Open
arsuhf wants to merge 1 commit into
openai:mainfrom
arsuhf:add-higgsfield-skills
Open

Add Higgsfield skills#404
arsuhf wants to merge 1 commit into
openai:mainfrom
arsuhf:add-higgsfield-skills

Conversation

@arsuhf
Copy link
Copy Markdown

@arsuhf arsuhf commented May 8, 2026

Summary

Adds four curated Higgsfield skills:

  • higgsfield-generate: MCP-based image/video generation, Marketing Studio, references, model selection, and media workflows.
  • higgsfield-soul-id: MCP-based Soul Character training and reuse.
  • higgsfield-product-photoshoot: CLI-based branded product photo generation through Higgsfield backend prompt enhancement.
  • higgsfield-marketplace-cards: CLI-based marketplace-ready product card generation.

Notes

higgsfield-generate and higgsfield-soul-id use the Higgsfield MCP server at https://mcp.higgsfield.ai/mcp.

higgsfield-product-photoshoot and higgsfield-marketplace-cards intentionally use the Higgsfield CLI because those backend enhancer workflows are currently exposed through CLI commands.

@arsuhf arsuhf requested a review from a team May 8, 2026 11:07
@arsuhf
Copy link
Copy Markdown
Author

arsuhf commented May 8, 2026

Hi! Thanks for taking a look.

This PR adds four Higgsfield skills under skills/.curated:

  • two MCP-based skills using https://mcp.higgsfield.ai/mcp
  • two CLI-based skills for backend enhancer workflows that are currently exposed through the Higgsfield CLI

I’ve validated the skills with quick_validate.py and smoke-tested the Higgsfield MCP integration through Codex (models_explore and a completed video generation job).

Happy to adjust scope, move anything to experimental, or split this PR if that makes review easier.

@RipperMercs RipperMercs mentioned this pull request May 9, 2026
5 tasks
@arsuhf
Copy link
Copy Markdown
Author

arsuhf commented May 12, 2026

@codex review

@chatgpt-codex-connector
Copy link
Copy Markdown

@arsuhf
Copy link
Copy Markdown
Author

arsuhf commented May 12, 2026

@codex review

Copy link
Copy Markdown

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 329b04171a

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

| `nano_banana_2` | Google | Character, cartoon, stylized, and reference-driven image work. |
| `soul_2` | Higgsfield | Aesthetic UGC, fashion editorial, lifestyle character generation; accepts Soul Character IDs. |
| `soul_cinematic` | Higgsfield | Cinematic stills with Soul identity support. |
| `soul_cast` | Higgsfield | Distinctive text-only personas. |
Copy link
Copy Markdown

Choose a reason for hiding this comment

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

P2 Badge Move soul_cast to the video model list

This catalog classifies soul_cast as an image model, but the official Higgsfield CLI model catalog lists soul_cast under video models (not image). In this skill, model tables are used to pick tool paths, so this misclassification can cause the agent to route soul_cast requests through generate_image instead of generate_video, leading to invalid calls or retries for users asking for Soul Cast outputs.

Useful? React with 👍 / 👎.

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