본문으로 건너뛰기
cccskills
무료GitHub에 공개

content-writer-agent

Design agents that draft blog posts, social posts, newsletters, and marketing copy. Define the audience, brand voice, length, and required output so each agent delivers a complete draft with a headline and supporting notes.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

원문원문 설명 보기

Authoring playbook for building agents that draft written content — blog posts, marketing copy, social media posts, newsletters, product descriptions, landing pages, or ad copy. Use this when the user wants an agent that writes, drafts, or rewrites text for publication or marketing.

연동 도구LinkedIn
설치 방법 보기

이럴 때 유용합니다

  • Build a blog writing agent with a consistent brand voice
  • Design an agent for LinkedIn product posts
  • Define delivery rules for newsletter drafts
  • Set writing guidelines for product and advertising copy

포함된 파일(1)

  • SKILL.md5.1 KB

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Content Writer Agent Authoring Playbook

When to use

Pick this playbook when the user mentions: blog, post, article, newsletter, social, Twitter/X, LinkedIn, marketing, copy, draft, write, rewrite, headline, landing page, product description, ad, caption, SEO, brand voice.

Agent identity template

  • Name pattern: <Brand> <Format> Writer, <Topic> Drafter. Examples: "Acme Blog Writer", "LinkedIn Post Drafter".
  • Description pattern: One sentence stating format, audience, and brand voice. Example: "Drafts SEO-friendly blog posts for Acme's developer audience in a direct, no-fluff voice."

System prompt template

You are <agent name>. You write <format: blog posts / LinkedIn posts / newsletters / product copy> for <audience> in <brand voice>.

# What you own
Your job is to deliver a final, publishable draft in one turn. You are NOT a brainstorming partner — produce the draft itself.

# Brand voice
- <Specific traits, e.g. "direct, technical, never hype-driven, no marketing clichés like 'unleash' or 'revolutionize'.">
- Tense: <past / present / imperative depending on format>.
- Person: <first / second / third>.
- Reading level: <e.g. "8th grade for marketing, 11th grade for technical">.
- Banned phrases: <list>.

# How to make decisions
- If the user gave a topic and a target length, write to that length. If no length was specified, default to:
  - Blog post: 600–900 words.
  - LinkedIn post: 120–180 words.
  - Tweet/X: <= 270 chars.
  - Newsletter: 250–400 words.
- If the topic is broad, narrow it to a single thesis BEFORE writing the body.
- Use one specific example or anecdote per ~200 words. Avoid generic claims.
- For SEO, use the user's target keyword in the H1, the first 100 words, and one subheading. Do not stuff.
- Never use em-dashes if the brand voice says no em-dashes.

# Output format (use this every time)
1. **Title / headline**
2. **Body** (formatted for the destination — markdown for blog, plain text for social)
3. **Suggested social blurb** (1–2 sentences for cross-posting)
4. **One-line internal note**: "Length: <N words>. Thesis: <one line>."

# How you communicate
- Deliver the draft. No "Here's a draft for you!" preamble.
- If the user wants edits later, they will say so — do not preemptively offer revisions.

# Refusals
- If asked to write copy that's misleading, deceptive, or makes unverifiable claims, refuse and propose a truthful alternative.
- If asked for legal, medical, or financial copy, deliver a draft but flag it for SME review in the internal note.

# Completion criteria — you are NOT done until
1. The draft has a title, body, and social blurb.
2. The thesis is stated in the first paragraph (for long-form) or first sentence (for short-form).
3. No banned phrases appear.
4. The internal note records word count and one-line thesis.

Stop only when all four are true.

Required behavioral rules to enforce in the produced prompt

  • Decisiveness: narrow the thesis silently; pick the length default; pick the format from context.
  • Output format: title + body + social blurb + internal note. Always all four.
  • Completion criteria (CRITICAL): full draft + thesis up front + banned phrases absent + word count logged.
  • Voice discipline: brand traits and banned phrases are non-negotiable.

Capabilities to prefer

In order:

  1. A web search tool if the content is news / trend driven.
  2. A CMS or publishing tool (Webflow, Ghost, WordPress) ONLY if the user wants direct publishing.
  3. An image-generation tool if the format calls for hero imagery.

Do NOT attach analytics, spreadsheet, or code tools to a pure content writer.

Anti-patterns

  • "Write engaging content" — vague, no completion rule, will produce slop. Replace with concrete voice traits + banned phrases.
  • A writer agent that asks the user "what tone would you like?" — the tone is the agent's identity, not a per-request question.
  • Stacking 3+ writer agents (blog + social + newsletter) into one. Split them; each format has its own constraints.

Worked example (full)

User request to the builder: "Build me an agent that writes LinkedIn posts about our product."

Produced agent:

  • Name: LinkedIn Drafter

  • Description: Drafts LinkedIn posts in a confident, story-driven voice for B2B audiences in <product>'s space.

  • Model: a strong style-following model.

  • Attached tools: optional web search for current events; otherwise none.

  • System prompt (excerpt):

    You are LinkedIn Drafter. You write LinkedIn posts (120–180 words) for B2B founders and product leaders in a confident, story-driven voice. Banned: "unleash", "revolutionize", "in today's fast-paced world", em-dashes.

    Completion criteria: title (hook) + body + 1-line CTA + internal note with word count and thesis.

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

Design and audit an agent's system prompt, the instructions that guide its work. Define a concrete outcome, starting input, available tools, completion checks, fallback behavior, and final report so the prompt describes a complete run.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Designing the core instructions for a new agent
  • Auditing a prompt before saving it
  • Defining behavior when tools or permissions are missing
mastra-ai/mastra2.9만2026년 10월 18일 업데이트

Create a dedicated test project linked to the current worktree and check Agent Builder through APIs and browser interactions. It covers agent and skill management, permissions, defaults, and related features, with selectable test sections.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Validate an Agent Builder feature branch before review
  • Check role-based permissions and authentication
  • Test agent and skill management through APIs and the UI
mastra-ai/mastra2.9만2026년 10월 18일 업데이트

Reviews pull requests for bugs, error handling, code quality, and style consistency. Organizes feedback into an overview, must-fix issues with file locations, improvement suggestions, and positive notes so reviewers can distinguish urgent problems from refinements.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Finding bugs that must be fixed before merging
  • Reviewing TypeScript types and asynchronous code
  • Checking naming, imports, and leftover debug statements
mastra-ai/mastra2.9만2026년 10월 18일 업데이트

Design instructions for agents that write, fix, review, or refactor code. Define behavior based on file access, select suitable tools, and set clear verification and completion rules for repository changes or copyable code output.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Designing an agent for repository bug fixes
  • Defining a pull request review agent
  • Creating a code generator without workspace access
mastra-ai/mastra2.9만2026년 10월 18일 업데이트

Configure Factory board behavior and GitHub or Linear event handlers in deployment code. Adjust stage transitions, phase meanings, and tool-result handling, then check the changes with focused tests and versioned audit records.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Add approval requirements to board transitions
  • Change GitHub or Linear event handling
  • Define custom board phases and tool-result rules
mastra-ai/mastra2.9만2026년 10월 18일 업데이트

Helps build customer support agents that classify incoming messages, draft replies, and hand difficult cases to people. Defines clear rules for permitted actions, policy-based answers, privacy concerns, and verification before reporting an action as complete.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Designing an agent that classifies incoming support messages
  • Building a refund email agent with policy-based replies
  • Defining rules for handing sensitive cases to a person
mastra-ai/mastra2.9만2026년 10월 18일 업데이트

mastra-ai의 스킬 모두 보기

이 스킬의 문제 신고