Skip to content
Back to skills

Customaize Agent

ASecurity

Verify skills work under pressure and resist rationalization using the RED-GREEN-REFACTOR cycle. Critical for discipline-enforcing skills. - Purpose - Test skill effectiveness with pressure scenarios - Output - Verification report with rationalization table ```bash /customaize-agent:test-skill ["skill path or name"] ```

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 25, 2026
ai-agentsgobashtestingcode-review

Security analysis

A100/100

Pro scans all 14 files and shows the line behind each finding

Scanned September 25, 2026

npx -y skills add Najeebullah3124/awesome-claude-universe --skill customaize-agent --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Customaize Agent?

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

Security grade badge for Customaize Agent
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/najeebullah3124-customaize-agent-awesome-claude-universe/badge)](https://www.skillsdirectory.com/skills/najeebullah3124-customaize-agent-awesome-claude-universe)

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
# /customaize-agent:test-skill - Skill Pressure Testing

Verify skills work under pressure and resist rationalization using the RED-GREEN-REFACTOR cycle. Critical for discipline-enforcing skills.

- Purpose - Test skill effectiveness with pressure scenarios
- Output - Verification report with rationalization table

```bash
/customaize-agent:test-skill ["skill path or name"]
```

## Usage Examples

```bash
# Test a TDD enforcement skill
> /customaize-agent:test-skill tdd

# Test a custom skill by path
> /customaize-agent:test-skill ~/.claude/skills/code-review/

# Start testing workflow
> /customaize-agent:test-skill
```

## Arguments

Optional path to skill being tested or skill name.

## How It Works

1. **RED Phase - Baseline Testing**: Run scenarios WITHOUT the skill
   - Create pressure scenarios (3+ combined pressures)
   - Document agent behavior and rationalizations verbatim
   - Identify patterns in failures

2. **GREEN Phase - Write Minimal Skill**: Address baseline failures
   - Write skill addressing specific observed rationalizations
   - Run same scenarios WITH skill
   - Verify agent now complies

3. **REFACTOR Phase - Close Loopholes**: Iterate until bulletproof
   - Identify NEW rationalizations from testing
   - Add explicit counters for each loophole
   - Build rationalization table
   - Create red flags list
   - Re-test until bulletproof

## Pressure Types

| Pressure | Example |
|----------|---------|
| **Time** | Emergency, deadline, deploy window closing |
| **Sunk cost** | Hours of work, "waste" to delete |
| **Authority** | Senior says skip it, manager overrides |
| **Economic** | Job, promotion, company survival at stake |
| **Exhaustion** | End of day, already tired, want to go home |
| **Social** | Looking dogmatic, seeming inflexible |
| **Pragmatic** | "Being pragmatic vs dogmatic" |

## Best Practices

- Combine 3+ pressures - Single pressure tests are too weak
- Document verbatim - Capture exact rationalizations, not summaries
- Iterate completely - Continue REFACTOR until no new rationalizations
- Use meta-testing - Ask agents how skill could have been clearer
- Test all skill types - Discipline-enforcing, technique, pattern, and reference skills need different tests

Files in this skill

  • README.md4.8 KB
  • agent-evaluation.md893 B
  • apply-anthropic-skill-best-practices.md2.6 KB
  • context-engineering.md1.2 KB
  • create-agent.md2.8 KB
  • create-command.md2.4 KB
  • create-hook.md4.6 KB
  • create-skill.md2.3 KB
  • create-workflow-command.md2.6 KB
  • prompt-engineering.md1.2 KB
  • test-prompt.md2.6 KB
  • test-skill.md2.2 KB
  • usage-examples.md13.6 KB

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…