Token导航 LogoToken导航TokenDH.com
待分类执行命令github未标认证来源可访问许可证需确认审计通过

fusion-skill-self-report-bug融合技能自报 bug

Agent Skill

fusion-skill-self-report-bug 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

1,285

周安装

52

GitHub Stars

公开资料未说明

下载量

404
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:fusion-skill-self-report-bug(融合技能自报 bug)
来源仓库:https://github.com/equinor/fusion-skills
仓库路径:skills/fusion-skill-self-report-bug
安装命令:
npx skills add https://github.com/equinor/fusion-skills --skill fusion-skill-self-report-bug
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/equinor/fusion-skills --skill fusion-skill-self-report-bug

简介

用于处理 GitHub 仓库协作信息与代码变更跟踪。

  • 适合查询 Issue、PR 状态或协助分支管理操作。
  • 可结合 token 权限控制对私有仓库的访问范围。
  • 写入类操作前应确认用户授权与仓库访问权限。
  • fusion-skill-self-report-bug 属于待分类类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Self-report Fusion Skill Bugs

When to use

Use this skill when a Fusion skill workflow fails and you need a consistent, triage-ready bug report draft. Treat this as the default failure handoff for any fusion-* skill execution.

Typical triggers:

  • "a fusion-* skill failed"
  • "this skill run failed"
  • "self-report this skill error"
  • "create a bug from this workflow failure"
  • "capture this automation failure for triage"

When not to use

Do not use this skill for:

  • Feature requests or roadmap planning
  • General implementation tasks without observed failure
  • Publishing GitHub changes without explicit user confirmation

Required inputs

Collect before drafting:

  • Target repository for the bug report (default: equinor/fusion-skills)
  • Failing command or workflow step
  • Environment context (OS/shell/runtime/tooling versions if available)
  • Observed output/error evidence
  • Reproduction signal (exact steps or best-effort notes)
  • Optional parent issue number for linking

Instructions

  1. Detect or confirm failure intent.

- If failure context is missing, ask for command used, expected behavior, actual behavior, and key error output.

  1. Capture failure context.

- Normalize context into: command, environment, observed error/output, reproducibility, and impact.

  1. Draft locally first.

- Write .tmp/BUG-skill-failure-<context>.md using assets/issue-templates/skill-workflow-failure-bug.md. - Include reproducible steps and explicit observed/expected behavior.

  1. Propose issue metadata.

- Recommend issue type: Bug. - Recommend labels for reliability/automation triage (for example bug, automation, reliability), then validate labels against the target repository before mutation. - Ask assignee preference (@me, specific user, or unassigned).

  1. Offer optional relationship linking.

- If parent issue is provided, prepare to link the new bug as a sub-issue after issue creation.

  1. Ask explicit publish confirmation.

- Do not run any GitHub mutation until the user explicitly confirms publish.

  1. Mutate only after confirmation.

- Create issue via MCP issue mutation. - Apply labels/assignee. - If parent issue was provided, link as sub-issue.

  1. If confirmation is not provided, stop after draft.

- Return status No GitHub state changes made.

Expected output

Return:

  • Draft bug report file path under .tmp/
  • Captured failure-context summary (command, environment, observed output)
  • Proposed title/body and reproduction steps
  • Proposed issue type + labels + assignee plan
  • Optional parent issue linking plan
  • Explicit status: Awaiting publish confirmation or No GitHub state changes made
  • Created issue URL/number only after confirmed mutation

Safety & constraints

Never:

  • Perform GitHub create/edit/comment/close actions without explicit confirmation
  • Claim failure evidence that was not provided or observed
  • Request or expose secrets/credentials

Always:

  • Keep draft-first behavior
  • Prefer minimal reproducible detail over long narrative
  • Validate label existence in the target repository before applying

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

36.68%
按下载量换算148

Claude

32.91%
按下载量换算133

Cursor

17.16%
按下载量换算69

Gemini CLI

10.07%
按下载量换算41

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/equinor/fusion-skills --skill fusion-skill-self-report-bug 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

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