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

paw-ps-research-to-briefPaw PS 研究简介

Agent Skill

paw-ps-research-to-brief 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

451

周安装

19

GitHub Stars

25

下载量

158
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:paw-ps-research-to-brief(Paw PS 研究简介)
来源仓库:https://github.com/pawbytes/skill-suites
仓库路径:skills/paw-ps-research-to-brief
安装命令:
npx skills add https://github.com/pawbytes/skill-suites --skill paw-ps-research-to-brief
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/pawbytes/skill-suites --skill paw-ps-research-to-brief

简介

paw-ps-research-to-brief 用于查找、检索和筛选相关信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中按任务场景定位内容。
  • 通过 npx skills add 命令从 pawbytes/skill-suites 仓库安装。
  • 使用前应核实权限、维护状态及是否执行网络请求或数据处理。
  • 建议结合原始 README 了解功能边界和调用方式。

SKILL.md

Research to Brief

Overview

A deterministic workflow that converts validated research outputs into a clean, structured product brief. This pipeline ensures no evidence or uncertainty is lost during synthesis.

Args: Accepts --headless / -H for non-interactive execution.

Output: Product brief document with preserved signals (evidence and uncertainties).


Pipeline

Stage 1: Gather Inputs

Load ./references/01-gather-inputs.md

Collect all relevant research files from the product workspace:

  • Discovery outputs (product-context.md)
  • Market intelligence (market-intelligence.md)
  • Audience intelligence (audience-intelligence.md)

Progression: Proceed when at least one input source is located.

Stage 2: Validate Completeness

Load ./references/02-validate-completeness.md

Assess whether inputs meet minimum thresholds for brief synthesis. Identify gaps and either proceed with available data or flag for additional research.

Progression: Proceed when minimum input requirements are met OR explicit decision to proceed with gaps documented.

Stage 3: Synthesize Brief

Load ./references/03-synthesize-brief.md

Create the structured product brief by consolidating validated inputs into a single document. Apply consistent formatting and ensure traceability to source materials.

Progression: Proceed when brief structure is complete with all required sections populated.

Stage 4: Preserve Signals

Load ./references/04-preserve-signals.md

Document key evidence sources and unresolved uncertainties. Create signal preservation notes that trace decisions back to research findings.

Progression: Complete when all signals are documented and appended to the brief or decision log.


On Activation

  1. Determine the product workspace root: {project-root}/.pawbytes/prodig-suites/
  2. Identify active product context from .pawbytes/prodig-suites/active-product.md if it exists
  3. If no active product is set and not in headless mode, prompt user to specify product slug
  4. Begin pipeline at Stage 1

Input Sources

SourceFile PathRequired
Discovery.pawbytes/prodig-suites/products/{product-slug}/product-context.mdPreferred
Market Research.pawbytes/prodig-suites/products/{product-slug}/market-intelligence.mdPreferred
Audience Research.pawbytes/prodig-suites/products/{product-slug}/audience-intelligence.mdPreferred

Minimum Requirement: At least one of the three input sources must be available.


Output Structure

.pawbytes/prodig-suites/products/{product-slug}/
├── product-brief.md              # Main brief document
├── signal-notes.md               # Evidence and uncertainty log
└── product-decisions.md          # Appended brief metadata (if exists)

Output Contract

Every execution delivers:

  • Action type: Brief synthesis from research inputs
  • Files saved:

- product-brief.md — Structured product brief - signal-notes.md — Evidence and uncertainty preservation

  • Recommendations: Next steps (e.g., validate with stakeholders, initiate design phase)
  • File saved to: .pawbytes/prodig-suites/products/{product-slug}/

Headless Mode

When invoked with --headless or -H:

  1. Attempt to read active product from .pawbytes/prodig-suites/active-product.md
  2. If not found, scan for most recently modified product workspace
  3. Execute all stages without prompts
  4. Document any gaps or assumptions in signal notes
  5. Exit with status code 0 on success, 1 if minimum inputs not met

Quality Gates

StageGate Condition
Gather InputsAt least 1 input file located
Validate CompletenessMinimum input threshold met OR documented decision to proceed
Synthesize BriefAll required brief sections populated
Preserve SignalsAll evidence and uncertainties documented

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

35.73%
按下载量换算56

Claude

27.69%
按下载量换算44

Cursor

19.4%
按下载量换算31

Gemini CLI

9.25%
按下载量换算15

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