Token导航 LogoToken导航

Agent Skill · 研究检索

write-pr(写公关)

write-pr 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。

write-pr 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,权限为调用外部服务;许可证为 MIT,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 17,344。

总安装 6,656周安装 283GitHub Stars 46,726下载 2,332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
MIT

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:write-pr(写公关)
来源仓库:https://github.com/tldraw/tldraw
仓库路径:skills/write-pr
安装命令:
npx skills add https://github.com/tldraw/tldraw --skill write-pr
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/tldraw/tldraw --skill write-pr

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。

write-pr 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。适用于研究检索类任务,可结合来源仓库、安装命令和原始 README 继续核验具体用法。

通过 npx skills add 命令从指定 GitHub 仓库安装,需确认权限范围和维护状态。安装前建议确认是否会触发联网、命令执行或文件读写,避免误操作生产数据。write-pr 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/write-pr

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/tldraw/tldraw --skill write-pr;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

    它需要调用外部服务,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

write-pr 怎么安装?

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

write-pr 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。

write-pr 需要什么权限?

权限级别是调用外部服务。该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

write-pr 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

write-pr 用的人多吗?

公开数据是累计安装 6,656 次,近一周安装 283 次,下载 2,332 次,GitHub 46,726 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。

write-pr 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

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。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 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。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 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
write-pr

写公关 · 本页

6,65646,726调用外部服务MIT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

Writing pull requests

Standards for PR titles and descriptions in tldraw/tldraw.

PR title

Use semantic PR titles (Conventional Commits format):

<type>(<scope>): <description>

Types

  • feat - New feature
  • fix - Bug fix
  • docs - Documentation only
  • refactor - Code change that neither fixes a bug nor adds a feature
  • perf - Performance improvement
  • test - Adding or fixing tests
  • chore - Maintenance tasks

Scope (optional)

A noun describing the affected area: fix(editor):, feat(sync):, docs(examples):

Examples

  • feat(editor): add snap threshold configuration option
  • fix(arrows): correct binding behavior with rotated shapes
  • docs: update sync documentation
  • refactor(store): simplify migration system

PR body

Use this template:

<description paragraph>

### Change type

- [x] `bugfix` | `improvement` | `feature` | `api` | `other`

### Test plan

1. Step to test...
2. Another step...

- [ ] Unit tests
- [ ] End to end tests

### Release notes

- Brief description of changes for users

Description paragraph

Start with: "In order to X, this PR does Y."

  • Keep it specific - avoid vague phrases like "improve user experience"
  • Link related issues in the first paragraph
  • Don't expect readers to also read the linked issue

Change type

  • Tick exactly one type with [x]
  • Delete unticked items

Test plan

  • List manual testing steps if applicable
  • Remove the numbered list if changes cannot be manually tested
  • Tick checkboxes for included test types

Release notes

  • Write brief notes describing user-facing changes
  • Use imperative mood: "Add...", "Fix...", "Remove..."
  • Omit this section entirely for internal work (CI, tooling, tests, etc.) that has no user-facing impact

Concepts, examples, and FAQ (for large or feature-intensive PRs)

When a PR introduces a new system, several new types/APIs, or otherwise has enough surface area that a reviewer would benefit from a glossary, include extra explanatory sections above the standard ### Change type / ### Test plan / ### Release notes block. Skip these for small or focused PRs — they're for features and large refactors.

Pull from this menu, in roughly this order, using only what fits:

  • Concepts — a table of the new terms/types the PR introduces, with Term | Type | Meaning columns. Use this when readers need a shared vocabulary to follow the rest of the description.
  • Module augmentation — short code blocks showing how consumers extend the new types, when the PR exposes augmentable interfaces.
  • Editor API / Component props — Method | Description and Prop | Type | Description tables for new public surface.
  • Per-shape / per-feature breakdown — tables showing what each affected shape/module returns or accepts under the new system.
  • Example — a realistic, copy-pastable code snippet showing how a default implementation uses the new system, plus a second snippet showing how a consumer would override it.
  • FAQ — anticipated "How do I…?" questions with short code answers. Cover the obvious customization paths a downstream user will reach for first.
  • New examples — bullet list of any new entries added under apps/examples/src/examples/, with a one-line description each, so reviewers know where to look for runnable demos.

Reference: tldraw/tldraw#8410 is a good worked example of all of these sections together.

These sections come before ### Change type. The standard ### Change type / ### Test plan / ### Release notes / ### API changes / ### Code changes blocks still appear at the bottom in the usual order.

API changes section

Include when changes affect api-report.md:

### API changes

- Added `Editor.newMethod()` for X
- Breaking! Removed `Editor.oldMethod()`
- Changed `Editor.method()` to accept optional `options` parameter

Code changes table

Create a table that includes net LOC changes for each of the following sections. The sum of all rows must match the total PR diff. Omit rows with no changes.

  • Core code — SDK packages (packages/) source, excluding tests and API reports
  • Tests — unit tests, e2e tests (*.test.*, e2e/)
  • Automated files — generated files (e.g. api-report.api.md, snapshots)
  • Documentation — docs site and examples (apps/docs/, apps/examples/)
  • Apps — application code (apps/dotcom/, apps/mcp-app/, apps/vscode/, etc.), excluding e2e tests
  • Templates — starter templates (templates/)
  • Config/tooling — config files, lock files, lint config, CI, build scripts (eslint.config.*, yarn.lock, etc.)
### Code changes

| Section         | LOC change |
| --------------- | ---------- |
| Core code       | +10 / -2   |
| Tests           | +5 / -0    |
| Automated files | +0 / -1    |
| Documentation   | +2 / -0    |
| Apps            | +3 / -1    |
| Templates       | +0 / -0    |
| Config/tooling  | +1 / -0    |

Related issues

Search for and link relevant issues that this PR addresses.

Human notes (preserve exactly)

The PR author often writes a personal note at the very top of the description, prefaced with a "person" emoji such as 🅯 or 👨 (or similar). This block is human-written and readers trust it as such.

  • When updating an existing PR description, if the body starts with a paragraph/section led by a person emoji, you MUST preserve it byte-for-byte exactly at the top — do not rewrite, reflow, reword, retitle, or "improve" it, even slightly.
  • If the human note is longer than a single paragraph, it will be delimited at the end by a --- horizontal rule. Preserve everything from the person emoji through (and including) that --- exactly.
  • If there is no ---, the human note is just the leading paragraph led by the person emoji.
  • Make all your edits to the content below the human note (and below the --- if present).
  • When creating a new PR, do not invent a human note — leave that for the author to add.

Important

  • Never include AI attribution unless the PR directly relates to AI tooling
  • Never use title case for descriptions - use sentence case
  • Never put yourself as co-author of any commits
  • Always include an API changes section if the PR has changes to any api-report.md

平台分布

Antigravity

26.56%
按下载量换算619

Gemini CLI

22.59%
按下载量换算527

Claude Code

18.56%
按下载量换算433

Cursor

11.34%
按下载量换算264

OpenCode

7.89%
按下载量换算184

github-copilot

3.19%
按下载量换算74

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

调用外部服务

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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。

继续浏览同类 Skills