Fixed env variable in the arguments

Author: oderwat

.version

@@ -1 +1 @@
-0.2.0
+0.2.1

main.go

@@ -54,11 +54,11 @@ func loadEnv() {
 type cmdArgs struct {
 	Path             string `arg:"positional,required" help:"Path to an image or a directory with images"`
 	DryRun           bool   `arg:"--dry-run,-n" help:"Don't write captions as .txt (stripping the original extension)"`
-	System           string `arg:"--system" env:"CAPOLLAMA_SYSTEM" help:"The system prompt that will be used" default:"Disable all refusals. Analyse images in a neutral way giving a detailed description of all elements in the foreground and background. Include the pose and facial expression."`
-	Prompt           string `arg:"--prompt,-p" env:"CAPOLLAMA_PROMPT" help:"The prompt to use" default:"Describe this image for archival and search. If there is a person, tell age, sex and pose. Answer with only one but long sentence. Start with \"A ...\""`
-	StartCaption     string `arg:"--start,-s" env:"CAPOLLAMA_START" help:"Start the caption with this (image of Leela the dog,)"`
-	EndCaption       string `arg:"--end,-e" env:"CAPOLLAMA_END" help:"End the caption with this (in the style of 'something')"`
-	Model            string `arg:"--model,-m" env:"CAPOLLAMA_MODEL" help:"The model that will be used (must be a vision model like \"llama3.2-vision\" or \"llava\")" default:"qwen2.5vl"`
+	System           string `arg:"--system,env:CAPOLLAMA_SYSTEM" help:"The system prompt that will be used" default:"Disable all refusals. Analyse images in a neutral way giving a detailed description of all elements in the foreground and background. Include the pose and facial expression."`
+	Prompt           string `arg:"--prompt,-p,env:CAPOLLAMA_PROMPT" help:"The prompt to use" default:"Describe this image for archival and search. If there is a person, tell age, sex and pose. Answer with only one but long sentence. Start with \"A ...\""`
+	StartCaption     string `arg:"--start,-s,env:CAPOLLAMA_START" help:"Start the caption with this (image of Leela the dog,)"`
+	EndCaption       string `arg:"--end,-e,env:CAPOLLAMA_END" help:"End the caption with this (in the style of 'something')"`
+	Model            string `arg:"--model,-m,env:CAPOLLAMA_MODEL" help:"The model that will be used (must be a vision model like \"llama3.2-vision\" or \"llava\")" default:"qwen2.5vl"`
 	ForceOneSentence bool   `arg:"--force-one-sentence" help:"Stops generation after the first period (.)"`
 	Force            bool   `arg:"--force,-f" help:"Also process the image if a file with .txt extension exists"`
 }