skill
lora-captioner-skill
Skill for AI agents to generate captions for your lora training
Set up
Skills are installed by placing their folder where your agent loads skills.
The download comes from github.com/sienadrayy/lora-captioner-skill itself, not from our server, so what you get is exactly what its author published.
Install instructions come from the README
README
synced Oct 23, 2026LoRA Captioner Skill for Codex
Create rich, high-detail captions for LoRA training images. The skill plans first, asks for your trigger word, and writes matching .txt files for each image.
What This Skill Does
- Prompts for the trigger word (never hardcodes it)
- Generates detailed, comma-tag captions
- Writes a .txt file with the same base name as each image
- Supports batch captioning without pauses
Install
Option A: Install with the Skill Installer (official)
Run this in your Codex terminal:
$skill-installer --repo sienadrayy/lora-captioner-skill --path . --name lora-captioner --method download
After installation, restart Codex to pick up the new skill.
Option B: Clone + Copy
Clone this repo and copy the folder into your Codex skills directory under local/lora-captioner.
Option C: Download ZIP + Copy
Download the repo ZIP, extract it, and copy the extracted folder into your Codex skills directory under local/lora-captioner.
Use It
In Codex, say something like:
Use the lora-captioner skill to caption all images in this folder.
The skill will:
- Plan first
- Ask for your trigger word
- Generate detailed captions and save .txt files
Repo Structure
- SKILL.md - skill definition
- agents/openai.yaml - agent configuration
Notes
- Captions are single-line, comma-separated tags
- The trigger word appears in every caption
- Backgrounds, lighting, pose, expression, and outfit are included when visible
SKILL.mdshould be saved as UTF-8 without BOM to avoid skill loader errors