An independent checker that tests whether a coding task really meets its stated completion conditions. It keeps the agent that wrote the code from judging its own work.

Use it after multi-step work, at the end of a goal cycle, or before submitting a pull request. It can check tests, type checks, linting, builds, changed-file boundaries, and other defined conditions.

It reduces the risk of declaring a task finished when tests, builds, or other requirements still fail. It reports each condition as passed, failed, or partly passed.

How chenklein26-maker/Harness-Starter configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to chenklein26-maker/Harness-Starter. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/chenklein26-maker/Harness-Starter/main/.claude/skills/verify-goal/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/chenklein26-maker/Harness-Starter

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/chenklein26-maker/harness-starter/verify-goal/github.svg)](https://agentmods.dev/skills/chenklein26-maker/harness-starter/verify-goal)

Your own site

<a href="https://agentmods.dev/skills/chenklein26-maker/harness-starter/verify-goal"><img src="https://agentmods.dev/badge/skills/chenklein26-maker/harness-starter/verify-goal/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/chenklein26-maker/harness-starter/verify-goal"><img src="https://agentmods.dev/badge/skills/chenklein26-maker/harness-starter/verify-goal.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 935 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00048	$0.00935
Opus 5	$0.00024	$0.00467
Sonnet 5	$0.00010	$0.00187
Haiku 4.5	$0.00005	$0.00093

Measured 9d ago against content hash a81e4ec4a816, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

verify-goal scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/verify-goal/SKILL.md · 101 lines

What it actually says

Verify Goal

核心理念

写代码的 Agent 不能给自己打分。

一个任务执行完毕时，所谓的"完成了"只是执行者的声称。verify-goal 作为独立的验证者，以完成条件为基准逐条验证，输出 PASS / FAIL / PARTIAL 结果。

触发条件

在以下场景中必须触发：

/goal 循环的每一轮结束后
多步骤任务执行完毕后，声称"完成了"之前
提交 PR 前的最终检查

不触发：

用户只是问问题，不需要执行
简单的文件操作（重命名、移动、删除）

工作流

Step 1：获取目标定义

从以下来源获取完成条件（按优先级）：

本次任务中用户或 AI 显式定义的完成条件（推荐格式见 CLAUDE.md Goal-Driven Execution）
对话上下文中可推导的隐含完成条件
任务本身的性质（如"修 bug" → 定义是"bug 复现步骤不再触发"）

Step 2：逐条验证

对每条完成条件，执行对应的验证命令并记录结果：

条件类型	验证方式	示例
测试通过	运行测试命令	`npm test -- --testPathPattern=auth`
类型检查	运行类型检查	`tsc --noEmit`
Lint	运行 linter	`npm run lint`
构建	运行 build	`npm run build`
边界条件	检查文件是否被误改	`git diff --name-only` 核对范围
自定义	执行用户定义的条件	按实际情况

Step 3：输出验证报告

使用以下固定格式输出：

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🔎 目标完成验证报告
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

目标：[本次任务的目标描述]

验证结果：✅ 全部通过 / ⚠️ 部分通过 / ❌ 未通过

逐项验证：
  1. [条件 A] ✅ PASS — [命令] → [输出摘要]
  2. [条件 B] ❌ FAIL — [命令] → [失败摘要]
  3. [条件 C] ✅ PASS — [命令] → [输出摘要]

─────────────────────────────────────
[全部通过]
  结论：目标已完成，可进入下一步。

[部分通过 / 未通过]
  失败项：
  - [条件 B] → [失败原因]
  建议：返回执行 Agent 修复后重新验证。
  降级：连续 3 次验证未通过 → 停止循环，汇报给人类。
─────────────────────────────────────

Step 4：判断下一步

根据验证结果决定后续动作：

验证结果	后续动作
✅ 全部通过	标记完成，更新状态文件，通知主流程继续
⚠️ 部分通过	将失败项返回执行 Agent 修复，修复后重新验证
❌ 全部未通过	停止循环，汇报给人类
连续 3 次 FAIL	Circuit Breaker 触发 — 停止循环，等待人类介入

与相关 skill 的关系

tech-review（事前）     verify-goal（事后）      acceptance（终验）
  方案查完了没有？   →   做完了条件满足了没？   →   整个流程走完了没？
  执行前触发           每轮 / 每次任务后触发       项目初始化结束时触发

三者形成完整链条：事前查方案 → 事后验结果 → 终验看全貌。

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

9d ago First seen · 101 lines · 48 tokens per session scan A a81e4ec4a816

Subscribe to this mod's changes

verify-goal is a skill published in the GitHub repository chenklein26-maker/Harness-Starter (119 stars, last pushed 2mo ago), licensed MIT. It adds 48 tokens to every session and 935 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Building agents Testing

research-engineer

An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.

davila7/claude-code-templates · 43 tokens

tika-eval-compare

Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".

apache/tika · 50 tokens

neuron-evaluation-engineer

Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…

neuron-core/neuron-ai · 77 tokens

Harness-Starter: Skill for Claude Code