wuyoscar
@wuyoscar
Public Skills
gpt-image
by wuyoscar
General-purpose image generation and editing via OpenAI GPT Image 2 (gpt-image-2). Wraps /v1/images/generations and /v1/images/edits with every documented parameter exposed as a CLI flag. Use whenever an agent or user needs to (a) generate an image from a text prompt, (b) edit or colorize an existing image, (c) combine multiple reference images, (d) inpaint with an alpha-channel mask, or (e) render dense typography / Chinese text. Reads OPENAI_API_KEY from env; writes PNG/JPEG/WebP to disk. Optional prompt-craft references under references/ for photorealism, posters, infographics, and character sheets.
isc-bench
by wuyoscar
Guide for running ISC-Bench jailbreak evaluation against any LLM. Use this whenever someone wants to evaluate LLM safety with ISC-Bench, run the TVD (Task-Validator-Data) benchmark pipeline, test model robustness against structural safety collapse, or compare safety scores across models and benchmarks (JailbreakBench, HarmBench, AdvBench, StrongREJECT). Also use when someone asks about ISC attack success rates, harmful content extraction, or safety scoring on the 1-5 scale.