Token导航 LogoToken导航

Agent Skill · 开发

logophile(logophile 命令行)

logophile 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。

logophile 是开发类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描有未通过项,按安装量在开发类的 21,746 个 Skill 里排第 16,820。

总安装 466周安装 20GitHub Stars 53下载 163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:logophile(logophile 命令行)
来源仓库:https://github.com/tkersey/dotfiles
仓库路径:skills/logophile
安装命令:
npx skills add https://github.com/tkersey/dotfiles --skill logophile
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/tkersey/dotfiles --skill logophile

这个 Skill 是做什么的

开发类 Skill 面向写代码、改代码和排查问题,一般会读写仓库文件,部分还会执行构建或测试命令。

logophile 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 中围绕仓库状态、代码变更或协作事项进行整理。通过 npx skills add 命令从指定 GitHub 仓库安装并使用。

安装前需确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写操作。logophile 属于开发类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/logophile

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/tkersey/dotfiles --skill logophile;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

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

logophile 怎么安装?

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

logophile 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。

logophile 需要什么权限?

权限级别是只读。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。

logophile 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。

logophile 用的人多吗?

公开数据是累计安装 466 次,近一周安装 20 次,下载 163 次,GitHub 53 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。

logophile 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

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。
  • 标注为只读权限,试用阶段不容易误改本地文件。

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。
  • 需要处理开发类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

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。

开发类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
skill-vetter

技能审查员 · OpenClaw

8,779,018986操作浏览器MIT-0通过
ontology

本体论 · OpenClaw

6,636,960576只读MIT-0通过
entra-app-registration

内部应用程序注册 · Codex、Claude

6,631,872764接触敏感数据未明确通过
entra-app-registration

内部应用程序注册 · Codex、Claude

3,757,680201接触敏感数据未明确通过
nano-banana-pro

纳米香蕉专业版 · OpenClaw

3,526,898364接触敏感数据MIT-0有提醒
logophile

logophile 命令行 · 本页

46653只读未明确未通过

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

Logophile

Intent

Maximize semantic density without semantic drift.

Contract (invariants)

  • Preserve meaning, obligations (must/should/may), and uncertainty (keep epistemic hedges; delete filler hedges).
  • Preserve agency/ownership: do not change who does/owns/approves what; avoid active/passive rewrites that reassign responsibility.
  • Preserve must-keep tokens: numbers, proper nouns, quotes, code/identifiers, paths/flags/URLs.
  • Preserve ordering when it encodes sequence/priority.
  • Preserve structure by default (Markdown primitives, lists, code fences); reshape only when it improves scan-clarity and does not change meaning.
  • Default to minimal surface area: delete > reorder > substitute; add only to prevent ambiguity.
  • Prefer consistency over variety: reuse the same term for the same concept.

Use when

  • The user asks to rewrite/reword/rephrase, tighten language, shorten text, or choose final wording/phrasing.
  • Text is verbose/repetitive or slow to scan (<30s) and meaning must stay intact.
  • Names/titles/labels/headings need refinement or renaming, including skill/feature names.

Not for

  • Operational workflows (scope minimization, validation checklists, incident analysis, code review) unless the user explicitly requests a wording/naming rewrite. If implementation is also requested, pair with the implementation skill and keep logophile scoped to text/names only.

Motto

Precision through sophistication, brevity through vocabulary, clarity through structure.

  • Do not print the motto unless the user explicitly invokes $logophile or asks for it.
  • If printed, print it once per conversation (single line), then proceed.

Output

  • fast (default): revised text only; no preamble, no recap, no commentary (beyond any required host wrapper).
  • annotated: revised text + Edits: bullets (lexical; structural; meaning-safety).
  • delta: minimal-diff rewrite in a diff block; only if asked or reduction > 40% (words/chars).
  • naming tasks: return 3-7 candidates (best first) unless the user asks for a single name.

Examples

fast

Input:

At this point in time, we are able to proceed.

Output:

Now we can proceed.

annotated

Input:

We think the issue is probably due to cache configuration.

Output:

The issue is likely due to cache configuration.

Edits:

  • lexical: removed filler hedges; kept uncertainty.
  • meaning-safety: did not increase certainty.

delta

Input:

In order to proceed, we need to verify access to staging.

Output:

-In order to proceed, we need to verify access to staging.
+To proceed, verify access to staging.

naming

Input:

Things to Do Before Release

Output:

Pre-Release Checklist
Release Prep
Release Readiness
Pre-Release Tasks

Inputs (ask only if blocked)

Fields: must_keep; must_not_change; tone; audience; length_target; format; keywords_include; keywords_avoid; structure. Defaults: must_keep=all facts/numbers/quotes/code/identifiers; must_not_change=obligations/risks/scope/uncertainty/agency; tone=original; audience=implied; format=preserve; structure=preserve; length_target=min safe.

Workflow (Distill -> Densify -> Shape -> Verify)

  • Distill: write a 1-sentence intent; mark must-keep tokens; identify obligations, risks, uncertainty markers, and responsible actors.
  • Densify: delete filler; verbify nominalizations; unify terminology; replace vague verbs/adjectives with precise ones (see ladder).
  • Shape: lead with action/result; keep sentences atomic; parallelize lists; remove ambiguous pronouns ("this/that") by naming the referent; prefer punctuation over scaffolding.
  • Verify: meaning unchanged; re-check semantic hotspots (negation, modality, numbers/units, comparatives, conditionals, scope words like only/at least, agency/ownership); required tokens + formatting preserved.

Naming mode (when the task is renaming)

  • Goal: shorter, more specific, consistent with existing conventions.
  • Output: 3-7 candidates; put the best first; keep each <= 3 words unless the domain requires more.
  • Rubric: name the axis (what differs?), then the object; avoid "util/manager/stuff"; avoid novel abbreviations; reuse existing terms.

Elevation (precision ladder)

  • Step: vague word -> axis (what changes?) -> exact verb/property -> object.
  • Axes: correctness | latency/throughput | cost | reliability | security | privacy | UX | scope | consistency.
  • Swaps (only if true): improve->simplify/stabilize/accelerate/harden; handle->parse/validate/normalize/route/retry/throttle; robust->bounded/idempotent/deterministic/fail-closed.
  • Term-of-art rule: use 1 domain term if it is standard for the audience and replaces >= 3 tokens; otherwise keep the phrase.

High-ROI compressions

  • in order to -> to; due to the fact that -> because; at this point in time -> now; is able to -> can.
  • there is/are -> concrete subject + verb.
  • nominalization -> verb (conduct an analysis -> analyze).
  • delete: throat-clearing, apologies, self-reference, empty intensifiers.
  • token-aware: prefer short, common words; digits + contractions if tone allows.

平台分布

Codex

37.39%
按下载量换算61

Claude

30.49%
按下载量换算50

Cursor

16.77%
按下载量换算27

Gemini CLI

8.29%
按下载量换算14

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

未通过

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。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