Skip to content
Agentmart
Submit a tool

eval-harness

克劳德代码会话的正式评估框架,实施评估驱动开发(EDD)原则

Skill by Luohaothu

IndexedMIT★ 26
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/luohaothu-eval-harness

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

克劳德代码会话的正式评估框架,实施评估驱动开发(EDD)原则

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/eval-harness

What's inside

Skill name
eval-harness
Description
克劳德代码会话的正式评估框架,实施评估驱动开发(EDD)原则
License
MIT
Instructions
256 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Eval Harness 技能

一个用于 Claude Code 会话的正式评估框架,实现了评估驱动开发 (EDD) 原则。

## 理念

评估驱动开发将评估视为 "AI 开发的单元测试":

* 在实现 **之前** 定义预期行为
* 在开发过程中持续运行评估
* 跟踪每次更改的回归情况
* 使用 pass@k 指标来衡量可靠性

## 评估类型

### 能力评估

测试 Claude 是否能完成之前无法完成的事情:

```markdown
[能力评估:功能名称]
任务:描述 Claude 应完成的工作
成功标准:
  - [ ] 标准 1
  - [ ] 标准 2
  - [ ] 标准 标准 3
预期输出:对预期结果的描述

```

### 回归评估

确保更改不会破坏现有功能:

```markdown
[回归评估:功能名称]
基线:SHA 或检查点名称
测试:
  - 现有测试-1:通过/失败
  - 现有测试-2:通过/失败
  - 现有测试-3:通过/失败
结果:X/Y 通过(之前为 Y/Y)

```

## 评分器类型

### 1. 基于代码的评分器

使用代码进行确定性检查:

```bash
# Check if file contains expected pattern
grep -q "export function handleAuth" src/auth.ts && echo "PASS" || echo "FAIL"

# Check if tests pass
npm test -- --testPathPattern="auth" && echo "PASS" || echo "FAIL"

# Check if build succeeds
npm run build && echo "PASS" || echo "FAIL"
```

### 2. 基于模型的评分器

使用 Claude 来评估开放式输出:

```markdown
[MODEL GRADER PROMPT]
评估以下代码变更:
1. 它是否解决了所述问题?
2. 它的结构是否良好?
3. 是否处理了边界情况?
4. 错误处理是否恰当?

评分:1-5 (1=差,5=优秀)
推理:[解释]

```

### 3. 人工评分器

标记为需要手动审查:

```markdown
[HUMAN REVIEW REQUIRED]
变更:对更改内容的描述
原因:为何需要人工审核
风险等级:低/中/高

```

## 指标

### pass@k

"k 次尝试中至少成功一次"

* pass@1:首次尝试成功率
* pass@3:3 次尝试内成功率
* 典型目标:pass@3 > 90%

### pass^k

"所有 k 次试验都成功"

* 更高的可靠性门槛
* pass^3:连续 3 次成功
* 用于关键路径

## 评估工作流程

### 1. 定义(编码前)

```markdown
## 评估定义:功能-xyz

### 能力评估
1. 可以创建新用户账户
2. 可以验证电子邮件格式

From Luohaothu/everything-codex, MIT.

More from Luohaothu/everything-codex

Signals

Stars
26
Forks
5
Open issues
0
Last commit (this folder)
Feb 5, 2026 (8 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from Luohaothu

Cite this page

eval-harness skill. Agent Mart. https://agentmart.dev/skills/luohaothu-eval-harness (last updated 2026-10-24)