Community编程与开发github.com

79109/agent-skills-test

综合矿山现场记录、监测变化和适用规范,比较可能原因、找出关键证据缺口,并提出能够区分解释的复核路径。适用于现场问题初步研判、原因排查、规范条款核对和处置思路讨论。

agent-skills-test 是什么?

agent-skills-test is a Claude Code agent skill that 综合矿山现场记录、监测变化和适用规范,比较可能原因、找出关键证据缺口,并提出能够区分解释的复核路径。适用于现场问题初步研判、原因排查、规范条款核对和处置思路讨论。.

兼容平台~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/79109/agent-skills-test/tree/HEAD/skill/mining-field-diagnosis

Installed? Explore more 编程与开发 skills: steipete/bluebubbles, steipete/eightctl, steipete/blucli · View all 6 →

在你喜欢的 AI 中提问

打开一个已预加载此 Agent Skill 的新对话。

文档

Mining Field Diagnosis

Focuses the investigation on "what could be the cause, how to judge it, and what to check next," transforming vague guesses about field issues into evidence-based investigation directions.

Core Capabilities

  • Compare competing explanations: Analyze multiple possible causes based on on-site conditions, retaining supporting and opposing evidence.
  • Map facts to standards: Check standard versions, scope of application, and actual evidence, highlighting non-compliance or missing conditions.
  • Prioritize key issues: Select the review actions that best distinguish between candidate causes, reducing aimless repetitive checks.

Getting Started

Read the phenomenon description, time and location, excavation or support conditions, monitoring records, equipment status, and measures already taken. Determine whether the user needs cause analysis, standard clause verification, or disposal strategy discussion; directly analyze parts with sufficient evidence, and ask targeted questions for critical gaps.

Immediate obvious dangers should be handled by on-site personnel and professionals following current emergency procedures, and disposal cannot be delayed while waiting for cause investigation or full standard verification.

Workflow

  1. First, distinguish between verified facts, personnel statements, and instrument clues.
  2. Compare candidate causes based on the problem mechanism, seeking observations that can distinguish explanations.
  3. When standard verification is needed, check the version, effective time, and applicable conditions, then match with on-site facts.
  4. Propose the most valuable review directions, preserving uncertainties in cause and compliance judgments.

Use Resources as Needed

  • For method selection, boundary judgment, or complex situations: Read references/methods.md.
  • For quickly organizing deliverables, copy and tailor assets/diagnosis-note.md; this is an optional starting point, not a fixed output format.

Deliverables

First, explain the judgments currently supported, then present the main candidate causes, key evidence, and priority review directions. When standards are involved, include necessary clause correspondences; do not force compliance checklists for tasks not involving standards.

Conclusions should be easy for professionals to review and do not replace on-site approval, acceptance, or resumption decisions. If the final cause is not determined, clearly express the uncertainty and the next steps that can advance the judgment.

Individual skills in this repo

This repo contains 4 individual skills — each has its own dedicated page.

相关技能