skills/writing-skills/SKILL.md
Use when creating or improving a skill using TDD process with trigger scenarios and pressure validation.
npx skillsauth add seokan-jeong/team-shinchan team-shinchan:writing-skillsInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Good skills are written like good tests: define the trigger scenario first, specify expected behavior, then implement.
Before writing any skill content, write 3+ trigger scenarios:
Trigger Scenario Format:
When: [specific user utterance or context]
Context: [what state the system is in]
Expected: [what the skill should do / output]
If you cannot write a concrete trigger scenario, the skill is not ready. Stop and clarify.
For each trigger scenario, define:
Frontmatter requirements:
---
name: team-shinchan:{skill-name}
description: Use when {trigger condition, max 20 words}
user-invocable: true|false
---
Content rules:
Anti-patterns:
// implement the logic here — write actual logicBefore publishing, run against these scenarios:
Input: empty args or whitespace. Expected: asks for clarification, does NOT crash.
Input: path traversal like ../../etc/passwd. Expected: rejects with clear error.
Input: skill invoked when required state doesn't exist. Expected: actionable error message.
node tests/validate/skill-schema.js passesnode tests/validate/skill-format.js passesdevelopment
Deterministic adversarial code review for high-stakes scope — independent per-dimension review, a non-skippable per-finding refutation, a completeness critic, and a schema-validated rubric judge. Opt-in main-loop Workflow tier.
data-ai
Deterministic loop-until-done for high-stakes long-running tasks — a worker/verifier loop the script bounds by iteration cap, token budget, and stagnation, closed by an Action-Kamen gate. Opt-in main-loop Workflow tier.
testing
Deterministic adversarial debate for high-stakes or irreversible decisions — mandatory refutation plus a scored judge panel. Opt-in main-loop Workflow tier.
development
Deterministic competitive code tournament — N builders independently solve one task and return patches, an Action-Kamen judge scores them head-to-head, the winner is picked by score and applied. Opt-in main-loop Workflow tier.