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

modifying-taxonomic-filter修改分类过滤器

Agent Skill

modifying-taxonomic-filter 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。

总安装

517

周安装

22

GitHub Stars

34,161

下载量

181
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:modifying-taxonomic-filter(修改分类过滤器)
来源仓库:https://github.com/posthog/posthog
仓库路径:skills/modifying-taxonomic-filter
安装命令:
npx skills add https://github.com/posthog/posthog --skill modifying-taxonomic-filter
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/posthog/posthog --skill modifying-taxonomic-filter

简介

用于查找、检索和筛选相关信息,支持关键词和任务场景定位。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中快速获取候选结果。
  • 可结合来源仓库和原始 README 继续核验具体用法。
  • 安装前建议确认权限范围和维护状态,避免触发联网或命令执行。
  • modifying-taxonomic-filter 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Modifying the TaxonomicFilter

The TaxonomicFilter is one of PostHog's most complex frontend components. It powers event/action/property/cohort selection across the entire app. Changes here have a high blast radius — always lock down behavior with tests before modifying.

Before you change anything

  1. Read the architecture — understand which logic and component you're touching. See references/architecture.md.
  2. Write RTL tests first — capture the current behavior before making changes. See references/testing-patterns.md.
  3. Check common pitfalls — avoid mistakes that have burned prior contributors. See references/common-pitfalls.md.
  4. Run the existing test suite to establish a green baseline: hogli test frontend/src/lib/components/TaxonomicFilter/

Architecture at a glance

TaxonomicFilter                    <- entry point, creates logic props
├── TaxonomicFilterSearchInput     <- keyboard events, search query
└── InfiniteSelectResults          <- tab pills + per-tab lists
    └── InfiniteList               <- virtualized list per group
        └── InfiniteListRow        <- individual result row

Three kea logics coordinate behavior:

LogicRoleKey file
taxonomicFilterLogicOrchestrator — tabs, search, group ordering, top matchestaxonomicFilterLogic.tsx (~1600 lines)
infiniteListLogicPer-tab list — search, pagination, property promotioninfiniteListLogic.ts
recentTaxonomicFiltersLogicRecent selections persisted to localStoragerecentTaxonomicFiltersLogic.ts

Each infiniteListLogic instance is keyed by listGroupType and connected to the parent taxonomicFilterLogic.

Key concepts

Group types

TaxonomicFilterGroupType is an enum with 40+ members (Events, Actions, PersonProperties, PageviewUrls, SuggestedFilters, etc.). Each group type maps to a TaxonomicFilterGroup configuration object defining its endpoint, search behavior, display name, and value extraction.

Tab ordering

Tabs are ordered by taxonomicGroupTypes prop order, but some groups get dynamically reordered. "Shortcut" groups (PageviewUrls, Screens, EmailAddresses, etc.) are promoted after SuggestedFilters when present. The groupAnalyticsTaxonomicGroupType selector normalizes group type names for analytics.

Search redistribution

redistributeTopMatches() distributes search results across groups:

  • Each group gets DEFAULT_SLOTS_PER_GROUP (5) slots
  • Empty groups donate their unused slots to REDISTRIBUTION_PRIORITY_GROUPS (CustomEvents, PageviewUrls, Screens)
  • Maximum MAX_TOP_MATCHES_PER_GROUP (10) results per group

Skeleton rows

While search results load, SKELETON_ROWS_PER_GROUP (3) placeholder rows appear. Use isSkeletonItem() to distinguish skeletons from real results.

Property promotion

infiniteListLogic promotes properties matching exact search terms to the top of results. PROMOTED_PROPERTIES_BY_SEARCH_TERM maps terms like 'url' to '$current_url' and 'email' to '$email'.

Testing workflow

Always write tests before modifying behavior. Two levels of testing:

Component tests (RTL)

Test what users see and interact with. Located in TaxonomicFilter.test.tsx.

const rendered = renderFilter({ taxonomicGroupTypes: [TaxonomicFilterGroupType.Events] })
await waitFor(() => expect(screen.getByTestId('taxonomic-filter-searchfield')).toBeInTheDocument())

Key patterns: renderFilter() helper, expectActiveTab() helper, userEvent for keyboard/click simulation. See references/testing-patterns.md for full details.

Logic tests (kea-test-utils)

Test state transitions and async behavior. Located in taxonomicFilterLogic.test.ts.

const logic = taxonomicFilterLogic(logicProps)
logic.mount()
await expectLogic(logic, () => logic.actions.setSearchQuery('pageview')).toMatchValues({ searchQuery: 'pageview' })

Key patterns: expectLogic(), toMatchValues, toDispatchActions, manual mounting of dependent infiniteListLogic instances. See references/testing-patterns.md for full details.

File reference

FileLinesPurpose
frontend/src/lib/components/TaxonomicFilter/TaxonomicFilter.tsx~240Entry component + search input
frontend/src/lib/components/TaxonomicFilter/taxonomicFilterLogic.tsx~1600Core orchestration logic
frontend/src/lib/components/TaxonomicFilter/infiniteListLogic.ts~500Per-tab list logic
frontend/src/lib/components/TaxonomicFilter/types.ts~280All type definitions
frontend/src/lib/components/TaxonomicFilter/InfiniteSelectResults.tsx~200Tab pills + list container
frontend/src/lib/components/TaxonomicFilter/InfiniteList.tsx~300Virtualized list
frontend/src/lib/components/TaxonomicFilter/TaxonomicFilter.test.tsx~660Component-level RTL tests
frontend/src/lib/components/TaxonomicFilter/taxonomicFilterLogic.test.ts~710Logic-level tests

Checklist for changes

  • Read the relevant architecture section
  • Write RTL tests capturing current behavior
  • Make your change
  • Verify all existing tests still pass
  • Add new tests for the changed behavior
  • Test keyboard navigation if you touched search or selection
  • Test with multiple group types if you touched tab ordering

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

36.83%
按下载量换算67

Claude

33.26%
按下载量换算60

Cursor

18.41%
按下载量换算33

Gemini CLI

8.79%
按下载量换算16

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