Skip to content
Back to skills

Tw Glaze

ASecurity

Run one explicit escalation pass using the original glaze prompt words verbatim. Use when prompts say `/tw-glaze`, when the first answer feels merely adequate, or when you want the exact original rhetoric preserved while pushing for a materially stronger direction.

  • 8 stars
  • 0 votes
  • 0 copies
  • 4 views
  • Added September 9, 2026
ai-agentsgo

Security analysis

A100/100

Scanned September 9, 2026

npx -y skills add SDiamante13/dotfiles --skill tw-glaze --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Tw Glaze?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Tw Glaze
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/sdiamante13-tw-glaze/badge)](https://www.skillsdirectory.com/skills/sdiamante13-tw-glaze)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: tw-glaze
description: "Run one explicit escalation pass using the original glaze prompt words verbatim. Use when prompts say `/tw-glaze`, when the first answer feels merely adequate, or when you want the exact original rhetoric preserved while pushing for a materially stronger direction."
---

# Glaze

## Verbatim Prompt

Use this text verbatim as the governing escalation text for the pass. Do not paraphrase, soften, summarize, or rewrite it.

```text
I think you can do much much much better than that! DIG DEEPER!!! RUMINATE HARDER!! BE BOLDER! MORE CREATIVE!
```

## Intent

Force one deeper pass before accepting an answer or plan, while preserving the original prompt language above exactly.

## Contract

- Start from the verbatim prompt text above.
- Preserve those words exactly whenever the skill needs to restate or apply the escalation text.
- Treat the current answer, plan, or instinct as insufficiently strong.
- Reject the first merely adequate move.
- Widen the search space once and look for a more accretive, more dominant direction.
- Prefer one strong thesis over a grab-bag of decent ideas.
- Keep the result grounded and operational, not hype-driven.
- If execution is requested, continue with the stronger plan after the reframing pass.

## Workflow

1. Name why the current answer underperforms.
2. Search for a better framing, invariant, or architecture.
3. Pick the highest-leverage move now available.
4. Explain why it dominates the obvious alternative.
5. Compress the result to the governing insight and next action.

## Output

- Default: `Why the obvious answer loses`, `Stronger move`, `Why it wins`, `Next action`.
- If no materially better move exists, say so explicitly instead of manufacturing novelty.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.