- Home
- Skills
- Image & Video
- Redthread — Red-Thread Portrait Collages
Works with the AI Agents you already use
Redthread — Red-Thread Portrait Collages
Turn one portrait into a poetic 3:4 collage with a glowing silhouette, textured paper, bilingual lettering and a red thread.
Free
About this skill
Redthread turns one user-supplied portrait into a finished 3:4 vertical, two-panel editorial collage.
The upper panel preserves the original scene and transforms the subject into a warm-white glowing silhouette. The lower panel reuses the same real subject cutout over the bundled fiber paper, recolored using a theme derived from the photo. Short English and Japanese phrases use the included DartsFont, and a wine-red colored-pencil thread connects corresponding subject anchors. When a little finger is clearly visible, it is the preferred anchor.
The workflow preserves the subject's face, pose, clothing, accessories and held objects. It delivers one inspected final PNG; cutouts, masks and component layers remain private working assets. You can also request revisions to the palette, wording, placement and connector.
Includes the skill instructions, visual style references, licensed font, fiber-paper texture, authored connector templates and Python rendering helpers. Requires a compatible agent with image compositing and pixel-preserving segmentation or background removal. Python with Pillow is the preferred helper runtime; a compatible existing compositor can be used instead.
Free and open source. Project code and paper texture use the MIT License. DartsFont is distributed under SIL Open Font License 1.1. Source and support: https://github.com/HuHana77/redthread
中文:将一张人物照片制作成红线双场景拼贴,上半为发光剪影,下半为真实人物与主题色纤维纸,配英日手写文字及酒红彩铅红线,仅交付最终 png。
Requires a user-supplied photo and a compatible host with pixel-preserving subject extraction and image compositing. The package does not include a segmentation model or a hosted image service. Output quality depends on the source photo and available extraction backend. The bundled font, paper and connector templates are required.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 28 days ago
- Free with an Agensi account
Requires
Permissions
Reads the user-provided source photo and bundled font, paper and connector templates. Writes private working layers and one final PNG. Shell access runs the bundled helper scripts. Network access is only needed if the user chooses a host-provided external image editing or segmentation service; no external account or API key is inherently required.