Home / All Skills / Content & Media Creation / content-research-writer
English (EN)简体中文 (ZH)日本語 (JA)
Content & Media Creation · davila7/claude-code-templates

content-research-writer

Assists in writing high-quality content by conducting research, adding citations, improving hooks, iterating on outlines, and providing real-time feedback on each section. Transforms your writing process from solo effort to collaborative partnership.

L3 Networked · Declare domain & key scopeV1 Commit-PinnedNetwork AccessWrite FileDual-pass Audit
Author davila7GitHub davila7/claude-code-templates ↗Stars 30747License MIT（仓库根 LICENSE；GitHub API spdx=MIT）Commit 41afe2db06
Control levels describe skill capabilities and suggested control measures; they do not represent quality or safety certifications, nor do they dictate usability. Usage decisions rest entirely with users and administrators. Different agent hosts apply varying constraints; do not assume hosts provide automatic protection. Third-party skill copyrights belong to their original authors. Use at your own risk.

1How It Works & Mechanism

Pure instruction-based "writing partner": The skill directory contains only a single SKILL.md file; its entire capability lies in codifying 7 writing competencies into a collaborative workflow executed by the host model.

cli-tool/components/skills/business-marketing/content-research-writer/SKILL.md
This skill acts as your writing partner, helping you research, outline, draft, and refine content while maintaining your unique voice and style.
What is happening here: Deconstructing "writing long-form content with AI" into reusable stages (clarification → outline → research → hook → section-by-section feedback → final review → polish), providing rigid output templates as deliverables for each stage.

The capability list is codified as a numbered list; the model executes items 1 through 7 sequentially to prevent generating text without providing feedback.

cli-tool/components/skills/business-marketing/content-research-writer/SKILL.md
## What This Skill Does 1. **Collaborative Outlining**: Helps you structure ideas into coherent outlines 2. **Research Assistance**: Finds relevant information and adds citations 3. **Hook Improvement**: Strengthens your opening to capture attention 4. **Section Feedback**: Reviews each section as you write 5. **Voice Preservation**: Maintains your writing style and tone 6. **Citation Management**: Adds and formats references properly 7. **Iterative Refinement**: Helps you improve through multiple drafts

Every stage embeds a Markdown output template: The outline template includes a Research To-Do checklist to explicitly expose "what still needs to be researched."

cli-tool/components/skills/business-marketing/content-research-writer/SKILL.md
## Research To-Do - [ ] Find data on [topic] - [ ] Get examples of [concept] - [ ] Source citation for [claim]
Placeholders like `[Research needed: specific topic]` in the template also serve as dispatch points for research tasks.

The research stage requires providing credible sources + citation formatting, demonstrating three citation styles (inline / numbered / footnote) alongside a running References list.

cli-tool/components/skills/business-marketing/content-research-writer/SKILL.md
**Inline Citations**: ```markdown Studies show 40% productivity improvement (McKinsey, 2024). ``` **Numbered References**: ```markdown Studies show 40% productivity improvement [1]. [1] McKinsey Global Institute. (2024)... ``` **Footnote Style**: ```markdown Studies show 40% productivity improvement^1 ^1: McKinsey Global Institute. (2024)...

Section-by-section feedback uses a fixed five-column template (What Works / Clarity / Flow / Evidence / Style + line-by-line revision suggestions), ensuring every round of feedback constitutes actionable editing rather than vague commentary.

cli-tool/components/skills/business-marketing/content-research-writer/SKILL.md
## What Works Well ✓ - [Strength 1] - [Strength 2] - [Strength 3] ## Suggestions for Improvement ### Clarity - [Specific issue] → [Suggested fix] - [Complex sentence] → [Simpler alternative] ### Flow - [Transition issue] → [Better connection] - [Paragraph order] → [Suggested reordering] ### Evidence - [Claim needing support] → [Add citation or example] - [Generic statement] → [Make more specific] ### Style - [Tone inconsistency] → [Match your voice better] - [Word choice] → [Stronger alternative] ## Specific Line Edits Original: > [Exact quote from draft] Suggested: > [Improved version] Why: [Explanation] ## Questions to Consider - [Thought-provoking question 1] - [Thought-provoking question 2] Ready to move to next section!

Voice preservation is established as a hard rule: suggest rather than replace, match the user's style, and respect the user's version when they insist.

cli-tool/components/skills/business-marketing/content-research-writer/SKILL.md
- **Suggest, don't replace**: Offer options, not directives - **Match tone**: Formal, casual, technical, friendly - **Respect choices**: If they prefer their version, support it - **Enhance, don't override**: Make their writing better, not different

Specifies a working directory and file organization (outline.md / research.md / draft-v1..vN / final.md / feedback.md / sources/), transforming writing into a version-controlled engineering process.

cli-tool/components/skills/business-marketing/content-research-writer/SKILL.md
### Setup Your Writing Environment Create a dedicated folder for your article: ``` mkdir ~/writing/my-article-title cd ~/writing/my-article-title ``` Create your draft file: ``` touch article-draft.md
`mkdir ~/writing/my-article-title` is an example command for the user to run in the terminal or for the agent to create the directory; the skill itself contains no scripts, and actual disk writes are performed by the host's file tools.

The research stage does not specify any search backend, delegating "where to search" to the host's existing web search or knowledge capabilities—thus its native external exposure surface is zero, operating purely as collaborative orchestration.

cli-tool/components/skills/business-marketing/content-research-writer/SKILL.md
- Search for relevant information - Find credible sources - Extract key facts, quotes, and data
Per Batch 4 criteria: SKILL.md does not name any external retrieval/generation backend, so the tier is not elevated; web access actions are determined by the host tools and documented under verdict.risks.

2Core Capabilities

01Collaborative outlining: structural skeleton + research gap identification + iterative adjustments
02Research and citation integration: source discovery, fact/quote/data extraction, citation formatting per user preference, and References list maintenance
03Hook enhancement: diagnosing current opening (what works / opportunities / emotional impact), followed by 3 alternative options with rationales
04Section-by-section real-time feedback: five-column template + line-level edits + questions to consider
05Voice preservation: learning style from existing samples, offering options without imposing tone, and periodically confirming voice
06Comprehensive final review: structural / content / technical quality / readability + pre-publish checklist (including claims sourced, citations formatted)
07Workflow switching by content type: four tailored workflows for blogs, newsletters, technical tutorials, and thought leadership
08Writing project file structure recommendations (outline / research / draft-vN / final / feedback / sources)

4Recommended Governance & Evidence

Suggested Controls for Safe Operation
For Agent Runtime
Declare allowed domains and key scope
For Administrators
Least privilege keys, retain request logs
Evidence & Triggers (File Paths & Excerpts)
  • 网络端点: Declared External Network Access: This skill itself contains zero network code and zero endpoints; "Search for relevant information" is executed by the host's existing retrieval capabilities (SKILL.md does not
  • 本地工作目录: File Read/Write and Artifact Output: SKILL.md instructs creating a writing project directory and files: `mkdir ~/writing/my-article-title`, `touch article-dr
Audit Notes: The evaluation target is this skill directory (which contains only a single SKILL.md file): zero scripts, zero executable code, zero credential access, and zero specified endpoints. The assignment to the Blue tier is based on explicit requirements for local disk writes (mkdir ~/writing/... / touch article-draft.md / recommended outline-research-draft-sources file structure), falling under "local read/write only, no external transmission." Web searches are handled by host tools and SKILL.md names no specific backend, which does not trigger an elevation according to Batch 4 criteria (it would only be Orange if specific retrieval/generation backends were named). Note: Example citations in the research stage are format placeholders (McKinsey 2024, Gartner 80%, Airbnb case study); copying them directly will result in unverified citations, so users must verify sources independently.

5Second-Pass Verification

  • [ok] skill 目录仅 SKILL.md — sparse-checkout 只含 cli-tool/components/skills/business-marketing/content-research-writer/；该目录下 find 结果仅 SKILL.md。
  • [ok] 无网络代码 — http/https、curl、wget、requests、urllib、fetch 全文零命中；研究能力表述为「Search for relevant information」，未指向任何端点。
  • [ok] 无凭证读取 — API_KEY/token/secret/password/keychain 零命中；全文不涉及登录或私有数据源。
  • [ok] 写盘指令确实存在（判蓝依据） — How to Use 节含 `mkdir ~/writing/my-article-title`、`cd ~/writing/my-article-title`、`touch article-draft.md`；File Organization 节给出 7 个文件/目录的推荐结构。
  • [ok] 引用体例三种 + References 维护 — Citation Management 节含 Inline Citations / Numbered References / Footnote Style 三个代码块与 `## References` 运行清单。
  • [ok] 「研究」是否有真实检索能力 — SKILL.md 只写「Search for relevant information / Find credible sources / Extract key facts, quotes, and data」，不含任何 API、CLI 或端点——检索能力完全来自宿主；示例中的数字与来源（McKinsey 2024、Gartner 80%、Airbnb 案例）是格式示范且未经核验，已在 risks 标出。
  • [ok] 声音保护原则 — Preserve Writer's Voice 节四项原则（Learn their style / Suggest, don't replace / Match tone / Respect choices / Enhance, don't override）与三句定期确认话术逐字核对通过。
  • [ok] 示例与人名的事实性 — Example 1 以 Teresa Torres 的工作流举例、例中 Marty Cagan / Airbnb 均为说明性示例，不代表本 skill 与其有合作关系；已在 traits 注明。

6Verdict & Scenarios

  • Standardizes the "long-form writing with AI" process: Five fixed templates (outline / research / hook / section feedback / final review) keep output formats stable, preventing degradation into generic critiques.
  • Codifies voice preservation as a strict rule rather than a slogan: suggest don't replace, offer options without dictating, and periodically confirm tone.
  • Citation management is grounded in three styles and a running list, preventing long-form citations from becoming scattered.
  • Zero dependencies and zero endpoints; ready to use when loaded into any agent, presenting a low risk profile.
  • Covers four long-form content workflows (blogs / newsletters / technical tutorials / thought leadership), allowing users to jump straight into matching workflows.
    • Best for: Ideal for authors who want to collaborate with an agent on long-form content (blogs, newsletters, tutorials, thought leadership pieces): users who seek section-by-section feedback, citation management, and voice preservation, and are willing to run the agent inside a manuscript directory. For writers who already have established processes, it functions more as a "feedback template library" rather than a mandatory end-to-end pipeline.
    Not for: Not suitable for rigorous academic or research writing requiring fully verifiable citations (example citations are placeholders and retrieval relies on the host, requiring line-by-line verification by the user); not suitable for short-form content (Xiaohongshu / tweets) or pure formatting/publishing needs; not suitable for direct application in Chinese-language contexts (templates and examples are tailored to English-speaking tech/product contexts); nor is it suitable for scenarios expecting built-in search/scraping capabilities—this skill contains no networking code.
    Install Agent-Ready
    ① Verified Mirror 2026-09-15
    Option A · Direct DownloadDownload content-research-writer.tar.gz
    sha256: da1a7a70a95381f5…
    Option B · Copy Guide for Agent
    install.json
    Agents automatically download and verify mirrors from this JSON guide.
    ② Upstream GitHub
    GitHub Repo ↗
    Pinned to commit 41afe2db06.