Model gemini-3.6-flashMeasured 8/20/2026v15 of 22 cases passed with skill · 19 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
A test case the skill fixed. Same model, same prompt, with and without the skill.
Prompt
Design an organic honey jar label visual. Generate a preliminary image draft, annotate honey drop texture and border layout edits on canvas, refine the graphic, and output structured action notes.
What was checked, without and with the skill
Checked
Without skill
With skill
Includes the label visual generation and texture refinement process.
Passed
Passed
Provides structured action notes detailing honey drop texture edits.
Use when: Use when the user asks to generate -> annotate -> refine -> repeat using generate_image + Visual Annotator Live Canvas.
Do not use when: Do not use when the request is asking for planning documents, high-level strategy, or non-executable discussion; use the relevant planning or design workflow instead.
Outputs: Outcome from Agentic Image Loop: task-specific result plus concrete action notes.
Success criteria: Returns concrete actions and decisions matching the requested task, with no fabricated tool-side behavior.
Trigger Examples
Positive
Use the agentic-image-loop skill for this request.
Help me with agentic image loop.
Use when the user asks to generate -> annotate -> refine -> repeat using generate_image + Visual Annotator Live Canvas.
Agentic Image Loop: provide an actionable result.
Negative
Do not use when the request is asking for planning documents, high-level strategy, or non-executable discussion; use the relevant planning or design workflow instead.
Do not use agentic-image-loop for unrelated requests.
This request is outside agentic image loop scope.
This is conceptual discussion only; no tool workflow is needed.
Runtime Prompt
Current runtime prompt length: 2117 characters.
Runtime prompt is defined directly in ../agentic-image-loop.json.