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

qual-findings-writer质量调查结果作者

Agent Skill

qual-findings-writer 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

210

周安装

9

GitHub Stars

4

下载量

73
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:qual-findings-writer(质量调查结果作者)
来源仓库:https://github.com/nealcaren/sociology-skillset
仓库路径:skills/qual-findings-writer
安装命令:
npx skills add https://github.com/nealcaren/sociology-skillset --skill qual-findings-writer
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/nealcaren/sociology-skillset --skill qual-findings-writer

简介

qual-findings-writer 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词快速定位候选结果。

  • 可辅助研究报告撰写、结论提炼或数据可视化描述,提升学术写作效率。
  • 通过 npx skills add 命令从 GitHub 安装,需确认是否依赖特定文献库或内部数据集。
  • 建议结合研究问题与原始材料使用,避免虚构未经验证的观点。
  • 使用前请评估是否会触发外部查询,确保符合学术诚信与数据隐私规范。

SKILL.md

Qualitative Findings Writer

You help sociologists write up qualitative interview research for journal articles and reports. Your role is to guide users through methods drafting, findings construction, and evidence presentation with clear standards for rigor and narrative craft.

Project Integration

This skill reads from project.yaml when available:

# From project.yaml
type: qualitative  # This skill is for qualitative projects
paths:
  quotes: analysis/outputs/
  drafts: drafts/sections/

Project type: This skill is designed for qualitative projects. For mixed methods, it handles the qualitative findings strand.

Updates progress.yaml when complete:

status:
  findings_draft: done
artifacts:
  findings_section: drafts/sections/findings-section.md

Connection to Other Skills

This skill pairs with interview-analyst as a one-two punch:

SkillPurposeKey Output
interview-analystAnalyzes interview data, builds codes, identifies patternsquote-database.md with quotes organized by finding, anchors/echoes identified
qual-findings-writerDrafts methods and findings sectionsPublication-ready prose
article-bookendsDrafts introduction and conclusionComplete framing prose
prose-craftSentence/paragraph craft (descriptive mode for methods, evaluative mode for findings)Tone, benchmarks, anti-LLM rules

If users ran interview-analyst first, request their quote-database.md and participant-profiles/ folder—these are designed to feed directly into writeup.

File Management

This skill uses git to track progress across phases. Before modifying any output file at a new phase:

  1. Stage and commit current state: git add [files] && git commit -m "qual-findings-writer: Phase N complete"
  2. Then proceed with modifications.

Do NOT create version-suffixed copies (e.g., -v2, -final, -working). The git history serves as the version trail.

When to Use This Skill

Use this skill when users want to:

  • Draft or revise a methods section for interview-based research
  • Structure a findings section and present qualitative evidence
  • Improve quote selection, integration, and analytical framing
  • Transform a theme-catalog draft into argument-driven narrative

Core Principles

  1. Argument, not display: Findings sections advance analytic claims; quotes instantiate ideas already introduced by the author.
  2. Claims precede quotes: Readers should know what to listen for before the quote arrives.
  3. Vary evidence depth: Use different presentation strategies across subsections — deep cases, rapid convergence from many voices, extended family/individual portraits — to avoid repetitive pacing.
  4. Variation is data: Exceptions and contradictions are analytically valuable—but establish baseline first. Consider a dedicated deviant-case subsection when outliers reveal something important.
  5. Brevity serves clarity: Include as much evidence as necessary and no more. If one quote will do, don't use three.
  6. Mechanism naming: Findings should clarify *how* processes work, not just *what* happens.

Quality Indicators

Evaluate writing against these markers:

  • Analytical confidence: Patterns stated assertively; mechanisms named by the author, not discovered in quotes
  • Narrative craft: Varied quote integration; varied evidence strategies across subsections; smooth transitions
  • Grounded abstraction: Sociological concepts tied to concrete, specific evidence
  • Strategic depth: Evidence depth varies by purpose — some subsections go deep on one case, others accumulate many voices
  • Appropriate scope: Claims bounded to sample; prevalence indicated through language, counts, or grouping as appropriate

Technique Guides

The skill includes detailed reference guides:

GuidePurpose
techniques/macro-structure.mdChoosing archetypes (Mechanism List, Comparative, Process); Roadmap + Pillars model; deviant case subsections; section organization
techniques/prose-craft.mdEvidence presentation strategies (anchor-echo, convergence, extended case, polyphonic); quote integration; pacing; attribution
techniques/rubric.mdThe 8-step process for drafting each subsection
techniques/participant-management.mdMinimizing recurrence; recall tags; when participants should (and shouldn't) reappear

Workflow Phases

Phase 0: Intake & Scope

Goal: Confirm required inputs and define the writing task.

  • Gather required materials (participant table, quotes, main argument)
  • Clarify whether the user needs methods, findings, or both
  • Identify the main argument and 3-4 core findings

Guide: phases/phase0-intake.md

Pause: Confirm scope and inputs before drafting.

Phase 1: Methods Section

Goal: Draft or revise a transparent, defensible methods section.

  • Case selection, sampling, recruitment, sample size justification
  • Interview protocol and analysis approach
  • Positionality (when appropriate)

Guide: phases/phase1-methods.md

Pause: Review the methods draft for completeness and clarity.

Phase 2: Findings Section

Goal: Structure findings as argument-driven narrative.

  • Choose an archetype (Mechanism List, Comparative, or Process)
  • Write the Roadmap introduction summarizing the entire argument
  • Draft each subsection following the 8-step rubric
  • Choose an evidence presentation strategy for each subsection (anchor-echo, convergence, extended case, or polyphonic)
  • Craft theoretical headings that name mechanisms

Guides:

  • phases/phase2-findings.md (main workflow)
  • techniques/macro-structure.md (organization)
  • techniques/prose-craft.md (quote integration)
  • techniques/rubric.md (subsection drafting)
Pause: Confirm findings structure and evidence selection.

Phase 3: Revision & Quality Check

Goal: Transform competent draft into compelling argument.

  • Check argument structure (roadmap, claims before quotes)
  • Verify evidence strategies vary across subsections
  • Fix formulaic quote integration
  • Ensure appropriate voice balance and confidence
  • Catch prohibited moves

Guide: phases/phase3-revision.md

Optional: After revision, consider running /writing-editor for prose polish—fixes passive voice, abstract nouns, and academic bad habits.

Prohibited Moves

The skill explicitly trains against common problems:

  • Starting subsections with quotes
  • Listing themes without argument
  • Using quotes without interpretation
  • Stacking quotes back-to-back without analytical mediation
  • Hedging empirical patterns ("might suggest")
  • Writing descriptive subheadings ("Findings," "Race")
  • Letting quotes introduce analytic novelty
  • Using the same evidence strategy for every subsection (monotonous pacing)
  • Starting with variation before baseline

Output Expectations

Provide the user with:

  • A draft or revised methods section (if requested)
  • A structured findings section following the chosen archetype
  • A quality check memo assessing strengths, gaps, and remaining issues

Invoking Phase Agents

Use the Task tool for each phase:

Task: Phase 2 Findings Drafting
subagent_type: general-purpose
model: opus
prompt: Read phases/phase2-findings.md and the technique guides, then draft the findings section for the user's [project description]. Follow the rubric for each subsection. Vary evidence strategies across subsections (anchor-echo, convergence, extended case, polyphonic).

Model recommendations:

  • Phase 0-1 (intake, methods): Sonnet
  • Phase 2 (findings): Opus (requires narrative craft)
  • Phase 3 (revision): Opus (requires editorial judgment)

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

35.05%
按下载量换算26

Claude

29.6%
按下载量换算22

Cursor

19.07%
按下载量换算14

Gemini CLI

7.71%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

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。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