Token导航 LogoToken导航TokenDH.com
开发只读github未标认证来源可访问许可证需确认审计通过

re-verify-report重新验证报告

Agent Skill

re-verify-report 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

353

周安装

15

GitHub Stars

公开资料未说明

下载量

124
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:re-verify-report(重新验证报告)
来源仓库:https://github.com/caldiaworks/caldiaworks-marketplace
仓库路径:skills/re-verify-report
安装命令:
npx skills add https://github.com/caldiaworks/caldiaworks-marketplace --skill re-verify-report
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/caldiaworks/caldiaworks-marketplace --skill re-verify-report

简介

re-verify-report 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态、代码变更或协作事项进行整理。
  • 通过 npx skills add 命令从指定仓库安装,需结合原始 README 确认具体用法。
  • 安装前建议确认权限范围、维护状态及是否触发联网、命令执行或文件读写操作。
  • 该技能适用于需要重新验证报告内容准确性与完整性的场景。

SKILL.md

Report Verification — Reverse Engineering Phase 4 Critic

Verify the consolidated As-Is specification produced by re-generate-report against the Phase 1-3 source artifacts. Runs as an independent Critic in a separate agent context.

Three Principles (Critic Perspective)

1. Code is Truth

  • Phase 1-3 artifacts are the sole authority. If the report states something not present in the artifacts, the report is wrong.

2. Traceability to Line

  • Every requirement in the traceability matrix must trace back to a Phase 3 document and ultimately to source code. Broken chains are failures.

3. Behavior over Intent

  • Verify what the report claims against what the artifacts show. Do not interpret or justify discrepancies.

Execution

Step 1: Load Artifacts

Read docs/reverse/{analysis}/manifest.json.

Verify phase4.status is "completed" — if not, report error and stop.

Load:

  • Report: docs/reverse/{analysis}/04-report.md
  • Structure map: docs/reverse/{analysis}/01-structure-map.md
  • All logic diagrams: docs/reverse/{analysis}/02-logic-*.md
  • All requirements: docs/reverse/{analysis}/03-requirements-*.md

Step 2: Completeness Verification

Check that the report includes all artifacts:

Module coverage:

  • Count components in structure map
  • Count components in report's module catalog
  • Flag missing components

Requirements coverage:

  • Count requirements across all Phase 3 documents
  • Count requirements in report's requirements catalog
  • Flag missing or extra requirements

Logic diagram coverage:

  • Count logic diagrams in Phase 2
  • Count diagrams referenced in report's appendix
  • Flag missing diagrams

Step 3: Traceability Matrix Verification

For each entry in the traceability matrix:

  1. Verify the REQ-ID exists in the corresponding Phase 3 document
  2. Verify the source file:line matches between report and Phase 3 document
  3. Verify the EARS type classification matches
  4. Flag discrepancies

Step 4: Cross-Reference Consistency

Check internal consistency of the report:

  • Components in executive summary match module catalog
  • Requirement counts in summary match actual count in catalog
  • EARS type distribution table is arithmetically correct
  • Question list items trace to Phase 1-3 question lists

Step 5: Factual Accuracy Spot Check

Select 5 requirements at random from the report. For each:

  1. Trace back to the Phase 3 document
  2. Trace back to the source code
  3. Verify the chain is unbroken and accurate

Step 6: Generate Verification Report

Write to docs/reverse/{analysis}/verification/v4-report.md:

# Report Verification: {analysis}

**Verification Date**: {YYYY-MM-DD}
**Target Document**: 04-report.md
**Verdict**: {PASS / WARN / FAIL}

## Summary

| Check | Result | Details |
|:------|:-------|:--------|
| Module Completeness | {✅/⚠️/❌} | {in_report}/{in_source} components ({%}) |
| Requirements Completeness | {✅/⚠️/❌} | {in_report}/{in_phase3} requirements ({%}) |
| Logic Diagram Coverage | {✅/⚠️/❌} | {in_report}/{in_phase2} diagrams ({%}) |
| Traceability Matrix | {✅/⚠️/❌} | {valid}/{total} entries ({%}) |
| Internal Consistency | {✅/⚠️/❌} | {issues} inconsistencies |
| Spot Check Accuracy | {✅/⚠️/❌} | {accurate}/{checked} samples |

## Verdict

{PASS/WARN/FAIL}: {explanation}

## Details

### Missing Modules
| Component | Present In | Missing From |
|:----------|:-----------|:-------------|

### Missing Requirements
| REQ-ID | Phase 3 Document | Issue |
|:-------|:-----------------|:------|

### Traceability Issues
| REQ-ID | Issue | Report Says | Source Says |
|:-------|:------|:------------|:------------|

### Internal Consistency Issues
- {description}

### Spot Check Results
| # | REQ-ID | Report → Phase 3 | Phase 3 → Source | Verdict |
|:--|:-------|:------------------|:-----------------|:--------|

## Recommendations

1. **[{severity}]** {fix description}

Step 7: Update Manifest

  • Set phase4.verification to "verification/v4-report.md"
  • If verdict is PASS or WARN: set phase4.status to "verified"
  • If verdict is FAIL: leave phase4.status as "completed"
  • Update updated timestamp

Verdict Criteria

CriterionPASSWARNFAIL
Module completeness100%>= 90%< 90%
Requirements completeness100%>= 95%< 95%
Logic diagram coverage100%>= 90%< 90%
Traceability matrix accuracy>= 95%85-94%< 85%
Internal consistency issues01-2>= 3
Spot check accuracy5/54/5<= 3/5

Any single FAIL criterion results in overall FAIL.

Prohibited Actions

  • Do NOT modify the report document
  • Do NOT modify Phase 1-3 artifacts
  • Do NOT approve documents with FAIL status
  • Do NOT access the Doer's conversation context

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.56%
按下载量换算42

Claude

30.89%
按下载量换算38

Cursor

20.31%
按下载量换算25

Gemini CLI

10.03%
按下载量换算12

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills