github:prime-skills/runcomfy-agent-skills · first seen 2026-10-30T20:36:17 · installs 385.3K · source · badge
Freshness 100 · Quality 75 · Efficacy 67
Skill Check baseline run sc_cca462c935ca at 2026-10-31T05:16:46+00:00: 3 auto-generated use cases x 3 models, graded by openai/gpt-4.1-mini. Cost $0.0152.
| Use case | google/gemini-2.5-flash | meta/llama-4-maverick | mistral/mistral-small |
|---|---|---|---|
| Generate a photorealistic portrait of a young woman with natural lighting and East Asian features. | PASS | PASS | PASS |
| Make an image of a poster with multi-line Japanese text and precise layout for an advertising campaign. | FAIL | PASS | PASS |
| Edit an existing image by swapping the background while preserving the identity of the main subject, using multiple input images if needed. | FAIL | PASS | FAIL |
| Pass rate | 1/3 | 3/3 | 2/3 |
Authors: run your own use cases with POST /api/skill-check (author-verified; never changes this public estimate). Health never affects weaving ranking.