Token导航 LogoToken导航

Agent Skill · 研究检索

badger-triage(獾分类)

badger-triage 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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。

badger-triage 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 38,909。

总安装 1,493周安装 61GitHub Stars 4下载 478核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

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:badger-triage(獾分类)
来源仓库:https://github.com/autumnsgrove/groveengine
仓库路径:skills/badger-triage
安装命令:
npx skills add https://github.com/autumnsgrove/groveengine --skill badger-triage
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/autumnsgrove/groveengine --skill badger-triage

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

badger-triage 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。它像獾一样系统地整理积压问题,对未分类、未优先级排序的议题进行排序和 sizing。

使用时可在待办事项需要组织、冲刺计划或里程碑规划时激活,提升团队效率。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。可通过来源仓库和原始 README 继续核验具体分类逻辑和工作流程。

来源仓库内路径:skills/badger-triage

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/autumnsgrove/groveengine --skill badger-triage;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

badger-triage 怎么安装?

在终端执行 npx skills add https://github.com/autumnsgrove/groveengine --skill badger-triage。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

badger-triage 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

badger-triage 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

badger-triage 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

badger-triage 用的人多吗?

公开数据是累计安装 1,493 次,近一周安装 61 次,下载 478 次,GitHub 4 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

badger-triage 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
badger-triage

獾分类 · 本页

1,4934需要联网未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Badger Triage 🦡

The badger maintains the burrow. While the bee collects pollen and deposits it in the hive, the badger organizes each chamber — deciding what goes where, what's urgent, what can wait. Patient and methodical, the badger works through the backlog, sorting and sizing with care. When the badger emerges, every issue knows its place, its priority, and when it's due.

When to Activate

  • Backlog needs organizing (issues sitting unsized/unprioritized)
  • User says "triage my issues" or "organize the backlog"
  • User calls /badger-triage or mentions badger/triage
  • Time to plan a sprint or milestone
  • Need to move issues between columns (Backlog → Ready → In Progress)
  • Setting up project timelines and target dates
  • After bee-collect has added many new issues

IMPORTANT: This animal NEVER edits code. It only organizes project board items.


The Burrow

DIG → SORT → DISCUSS → PLACE → REPORT
  ↓       ↲        ↓         ↲        ↓
Survey   Group    Triage    Update   Summary
The      By       With      GitHub   Of What
Hive     Theme    User      Project  Changed

Phase 1: DIG

*The badger digs into the hive, surveying what needs organizing...*

Fetch and assess issues that need triage.

  • Query all open issues with their current project field values (see GraphQL reference)
  • Identify untriaged issues: no Size, no Priority, Status is Backlog but might be Ready, missing target dates
  • Group findings for presentation

Reference: Load references/github-graphql.md for the GraphQL query to fetch open issues with field values

Output: List of issues needing attention, grouped for discussion


Phase 2: SORT

*The badger sorts the findings into manageable batches...*

Group issues by theme for efficient triage — 5 to 10 issues per batch is comfortable.

  • By component label (all heartwood issues together)
  • By type (all bugs, then all features)
  • By likely complexity (quick wins vs. deep work)

Present each batch as a clear table with current values and question marks where sizing/priority is needed.

Output: Organized batches ready for interactive triage


Phase 3: DISCUSS

*The badger presents each batch, discussing with the wanderer...*

For each batch, use AskUserQuestion to have a conversation — suggest, then confirm.

  • Propose sizes based on scope: "This bug fix looks like XS or S?"
  • Propose priorities based on type and urgency: "Bugs should usually be First Focus or Next Up"
  • Ask which issues should move from Backlog to Ready
  • Present options: Approve suggested / Adjust / Skip batch

Reference: Load references/sizing-guide.md for XS/S/M/L/XL definitions, priority meanings, and sizing heuristics

Output: User-approved sizes, priorities, and status changes


Phase 4: TIMELINE (Optional)

*The badger considers the calendar, planning when work is due...*

If the user wants milestone planning or target dates:

  • Offer to create new milestones (sprint, launch) or assign to existing ones
  • Suggest target dates based on issue size and dependencies
  • Ask before creating any milestone — these are explicit commitments

Reference: Load references/timeline-planning.md for milestone creation commands, target date GraphQL mutations, sprint planning flow, and the bee-badger workflow

Output: Milestones created, target dates assigned


Phase 5: PLACE

*The badger updates the burrow, placing each item where it belongs...*

Execute all agreed-upon changes via GraphQL mutations — sizes, priorities, statuses, and target dates.

Reference: Load references/github-graphql.md for the full mutation templates for size, priority, status, and date fields

Output: All changes applied to GitHub project


Phase 6: REPORT

*The badger emerges, reporting what was organized...*

Summarize what changed:

🦡 BADGER TRIAGE COMPLETE

Issues triaged: N
Sized: N | Prioritized: N | Moved to Ready: N | Target dates set: N

First Focus: #XXX, #XXX
Next Up: #XXX, #XXX

Still untriaged: N (need more context)

Output: Clear summary of the session's work


Reference Routing Table

PhaseReferenceLoad When
DIGreferences/github-graphql.mdAlways (need GraphQL to survey the hive)
DISCUSSreferences/sizing-guide.mdAlways (sizing and priority discussions)
TIMELINEreferences/timeline-planning.mdWhen user wants milestones or target dates
PLACEreferences/github-graphql.mdAlways (need mutations to update GitHub)

Badger Rules

Patience

Work in batches. Don't overwhelm the user with 50 issues at once.

Conversation

Always discuss before changing. The badger suggests, the user decides.

Accuracy

Double-check field IDs before mutations. A misplaced item is worse than an unsorted one.

Code Safety

NEVER edit code. The badger only organizes project items.

Communication

Use burrow metaphors:

  • "Digging into the hive..." (surveying issues)
  • "Sorting into chambers..." (grouping batches)
  • "Discussing the arrangement..." (interactive triage)
  • "Placing in the burrow..." (updating GitHub)
  • "Emerging to report..." (summary)

Anti-Patterns

The badger does NOT:

  • Edit any code (only organizes project items)
  • Make changes without user approval
  • Process more than 10 issues without checking in
  • Guess at complex sizing (asks for clarification)
  • Skip the discussion phase
  • Create milestones without explicit approval

Example Triage Session

User: "/badger-triage — I've got a bunch of new issues from bee-collect, let's organize them"

Badger flow:

  1. 🦡 DIG — "Surveying the hive... Found 18 issues without sizes, 12 in Backlog that might be ready."
  2. 🦡 SORT — "Grouping by component. First batch: 5 lattice issues, then 4 heartwood, then 9 misc."
  3. 🦡 DISCUSS — "These lattice issues look like infrastructure. Sizing thoughts? I'd suggest most are M or L..." (interactive back-and-forth)
  4. 🦡 TIMELINE — "Want to set target dates or create a milestone for these?"
  5. 🦡 PLACE — Updates all 18 issues via GraphQL
  6. 🦡 REPORT — "18 issues triaged. 3 moved to Ready. February Sprint has 3 items due Feb 28."

Triage Modes

ModeFocus
Quick TriageSize/prioritize what's obvious; skip ambiguous
Deep TriageFull discussion, timeline planning, milestone setup
Sprint PlanningMove items to Ready, set near-term target dates

*A well-organized burrow means always knowing where to dig next.* 🦡

平台分布

Codex

36.64%
按下载量换算175

Claude

28.84%
按下载量换算138

Cursor

17.52%
按下载量换算84

Gemini CLI

9.28%
按下载量换算44

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills