momentic-result-classification
Classify Momentic Test Run Failures
Test failures can hide their true cause behind later errors or retries. This skill investigates Momentic run evidence and produces a precise classification.
使用我的智能体安装
将此请求复制给你的智能体，其中包含权威 Skill 页面和清单。
Review the Skillstore skill "momentic-result-classification" from https://skillstore.io/skills/momentic-ai-momentic-result-classification.md and its manifest at https://skillstore.io/api/skills/momentic-ai-momentic-result-classification/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.你的智能体仍应展示其计划，并按安全策略请求必要确认。
Agent 可读资源
当 AI Agent、爬虫或脚本需要干净上下文，而不是读取完整页面时，请使用这些链接。
测试它
正在使用“momentic-result-classification”。 Classify a run where a row action succeeded, but the expected editor never opened.
预期结果:
Reasoning: The row action was the earliest divergence. The screenshot stayed on the parent page, so the later title assertion checked unrelated text. Category: BUG. Recoverable: NON_RECOVERABLE. Confidence: high.
正在使用“momentic-result-classification”。 Review a run where a broad text assertion passed before typing occurred in the wrong field.
预期结果:
Reasoning: The assertion did not prove the intended page state. A more specific assertion and target can prevent recurrence. Category: TEST_AUTHORSHIP. Recoverable: RECOVERABLE. Confidence: high.
正在使用“momentic-result-classification”。 Explain a run that failed because a required upload file was unavailable.
预期结果:
Reasoning: The required file was missing before the picker action, and the test cannot create it. Category: TEST_SETUP. Recoverable: NON_RECOVERABLE. Confidence: high.
安全审计
安全All 58 static findings are false positives caused by Markdown backticks, code fences, ordinary test-analysis prose, and a documented Momentic run-link template. The skill uses expected Momentic MCP tools, and no command execution, system reconnaissance, prompt injection, or unauthorized network behavior is present.
风险因素
⚙️ 外部命令 (44)
🌐 网络访问 (1)
分享与引用此报告
分享版本化评估报告、中性徽章、嵌入卡片和引用。Skillstore 只报告证据，不判断该 Skill 是否安全。
复制报告链接
https://skillstore.io/skills/momentic-ai-momentic-result-classification/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/momentic-ai-momentic-result-classification?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/momentic-ai-momentic-result-classification?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/momentic-ai-momentic-result-classification/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/momentic-ai-momentic-result-classification.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>学术引用 (APA · BibTeX · CFF)
APA 引用
momentic-ai. (2026). momentic-result-classification security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/momentic-ai-momentic-result-classification/audits/1BibTeX 引用
@techreport{momentic-ai-momentic-ai-momentic-result-classification-2026,
author = {momentic-ai},
title = {momentic-result-classification security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/momentic-ai-momentic-result-classification/audits/1},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "momentic-result-classification security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "momentic-ai"
date-released: "2026-08-22"
url: "https://skillstore.io/skills/momentic-ai-momentic-result-classification/audits/1"
identifiers:
- type: other
value: "skillstore:momentic-ai-momentic-result-classification:audit:1"
description: "Skillstore immutable audit report identifier"
Skillstore 评分
评分依据 证据置信度: 中你能构建什么
Triage a failed test run
Find the earliest divergence and classify the failure using direct run evidence.
Investigate intermittent failures
Compare attempts and past runs to distinguish consistent defects from temporary conditions.
Review release test failures
Summarize root causes, recoverability, and confidence for failed end-to-end checks.
试试这些提示
Investigate Momentic run [run ID]. Identify the earliest divergence, classify the failure, determine recoverability, and explain the evidence.
Explain why Momentic run [run ID] failed. Trace backward from the final error and cite the screenshots, URLs, or step results that prove the cause.
Compare Momentic run [current run ID] with relevant earlier runs on the same branch. Determine what changed and provide the formal classification.
Analyze every attempt and recovered child step in Momentic run [run ID]. Identify hidden partial failures, broken postconditions, category, recoverability, and confidence.
最佳实践
- Inspect the current run before using historical results.
- Verify causal claims with screenshots, step results, URLs, DOM data, or traces.
- Match past runs by branch and test version whenever possible.
避免
- Do not classify only from the final failing assertion.
- Do not label a single failure as flaky without comparable evidence.
- Do not infer overlays, page states, or application changes without direct evidence.
常见问题
What data does this skill need?
Can it compare previous runs?
Does it automatically save the classification?
Can it identify flaky tests?
What categories can it return?
Can it fix the failed test?
开发者详情
作者
momentic-ai许可证
MIT
Skillstore 修订
r1
版本提示
作者未声明版本。
引用
a740049399b99f93094598d14138485f301898d6
维护新鲜度
2026/8/22
使用量
0 次下载 · 0 次浏览
文件结构
📄 SKILL.md