Agent skills for ChatGPT

Manage local skills - list, add, remove, search, edit, setup wizard

Write MiniMax H3 Base and Ref2VA prompts.

Automatically select the optimal Claude model based on task complexity to reduce costs while maintaining quality.

Design kinetic type and motion-graphics beats.

Apply targeted negatives, not kitchen-sink bans.

Use neon practicals as motivated color keys.

Direct creature performance and non-human body.

Analyze options chains, implied volatility, Greeks, and strategy payoffs using EODHD Marketplace options data.

Direct micro-expressions and eye-acting beats.

Analyze portfolio risk — technical indicators, sentiment shifts, insider alerts, ESG scores, and volatility metrics.

Order positives: subject, light, lens, grade.

Move an idea up the fidelity ladder (content-first → wireframe → low-fi → high-fi → code) with a validation plan at each level, plus…

Defines the quantitative skill-optimization model: its run directory and state schema, four evaluation laws, verification-signal policy…

Improves an existing skill by running it as a training loop: evaluates it on real tasks against a mechanical verification signal, proposes…

Reviews a skill by first identifying whether its parts provide guidance, coordinate a workflow, or perform a task, then applies only the…

Read LTV:CAC against the 3:1 line and contribution margin after the real costs, so a profitability problem is diagnosed correctly.

Use this skill before publishing or sharing an agent-skill bundle, generated skill fixture, release candidate, or public runbook.

Upgrade an existing website or app to premium quality without breaking functionality — audit the current design, identify generic/AI…

Pick aspect, resolution, and delivery crop.

Use when the user is unsure which skill to use or where to start.

PRIORITY OVERRIDE protocol. Activates when trauma/crisis/vulnerability detected. Stops all frameworks, deploys presence-only mode. Always…