View and override LLM prompt templates used for media description (image/video/document) and gating decisions.
From omninpx claudepluginhub automagik-dev/omni --plugin omniThis skill is limited to using the following tools:
Designs and optimizes AI agent action spaces, tool definitions, observation formats, error recovery, and context for higher task completion rates.
omni prompts list --json
omni prompts get image --json
omni prompts get video --json
omni prompts get document --json
omni prompts get gate --json
# Set inline
omni prompts set image "Describe this image in detail, including any text visible." --reason "More detail needed" --json
# Set from a file using stdin (for multiline prompts)
omni prompts set document < /path/to/prompt.txt --json
# Set gate prompt
omni prompts set gate "Reply YES if the message is a support request, NO otherwise." --json
omni prompts reset image --json
omni prompts reset video --json
omni prompts reset document --json
omni prompts reset gate --json
image, video, document, gate.reset clears the override and reverts to the built-in code default.set reads from stdin if no inline value is provided, enabling multiline prompt editing.--reason documents why the prompt was changed (stored for audit purposes).