Skip to content
Back to skills

Verdict

ASecurity

Use to derive a deterministic reject/pass verdict from a draft@1 artifact's open-defect count, emitting a verdict@1 artifact for a branch to route on. Single tool. Reusable standalone.

  • 209 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 22, 2026
ai-agentspython

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned September 22, 2026

npx -y skills add s0912758806p/agentic-sop-to-work --skill verdict --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Verdict?

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

Security grade badge for Verdict
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/s0912758806p-verdict/badge)](https://www.skillsdirectory.com/skills/s0912758806p-verdict)

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: verdict
description: Use to derive a deterministic reject/pass verdict from a draft@1 artifact's open-defect count, emitting a verdict@1 artifact for a branch to route on. Single tool. Reusable standalone.
---

# Skill — verdict(判定,退回邊的判定來源)

圖示範流程的第 2 步。判定**完全確定性**:由上游 `defects` 導出,不由模型決定;
編排層再用 branch predicate 讀 `data.verdict` 決定往前或走退回邊——控制流始終在程式裡。

## 綁定的單一工具
- `python3`(標準庫)。**本 skill 只此一個工具。**

## 依賴
- `python` >= 3.8

## 參數化(無硬編碼)
- 輸入:`--in <draft artifact.json>`;輸出:`--out <verdict artifact.json>`

## 介面
- **輸入 artifact**:`draft@1`(讀 `data.defects`、`data.round`)。
- **輸出 artifact**:`verdict@1` — `data{verdict: "reject"|"pass", defects, round}`,透傳 `trace`。

## fail-loud 邊界
上游沒有 `defects` 欄位 → **exit 3**,不臆測判定(缺資料不得猜結論)。

## 執行
`python3 skills/verdict/tool.py --in <$RUN/draft.json> --out <$RUN/verdict.json>`

## 獨立重用
`skills/verdict/` + `lib/kit.py` 可單獨抽出;只要上游給 `draft@1` 即可運作。

Files in this skill

  • SKILL.md1.2 KB
  • tool.py1.3 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…

Newsletter

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