AGENT SKILL

platform-content

Platform-native content packs: Twitter/X threads, LinkedIn posts, YouTube and TikTok scripts, Instagram captions, newsletter issues, blog articles, podcast outlines, 30-day calendars and repurposing chains, each with hook, body and call to action, built from a brief about your platform, niche, audience and goal. Use whenever someone asks for social or audience content: write me a thread about X, a LinkedIn post that brings leads, turn this into ten posts, plan next month of content, a YouTube script people actually finish, or they bring a blog, newsletter or podcast idea that needs an asset per platform, even if they never say content strategy. Not for copy that sells a product (use sales-copy), rewriting text the user already wrote (use wordsmith), SEO cluster and cannibalization planning (use keyword-guard), or career documents (use career-track).

License 许可
MIT
最近更新
2026-11-14
Stars
★ 0
Forks
0

01 前言 Front Matter

name
platform-content
version
1.0.0
description
Platform-native content packs: Twitter/X threads, LinkedIn posts, YouTube and TikTok scripts, Instagram captions, newsletter issues, blog articles, podcast outlines, 30-day calendars and repurposing chains, each with hook, body and call to action, built from a brief about your platform, niche, audience and goal. Use whenever someone asks for social or audience content: write me a thread about X, a LinkedIn post that brings leads, turn this into ten posts, plan next month of content, a YouTube script people actually finish, or they bring a blog, newsletter or podcast idea that needs an asset per platform, even if they never say content strategy. Not for copy that sells a product (use sales-copy), rewriting text the user already wrote (use wordsmith), SEO cluster and cannibalization planning (use keyword-guard), or career documents (use career-track).

02 资源文件 2

03 SKILL.md 正文

Platform Content

Overview

Produces finished, platform-native assets (X threads, LinkedIn posts, YouTube and short-form scripts, Instagram captions, newsletter issues, blog articles, podcast outlines), plus a 30-day calendar and repurposing chains, from a brief that names the platform, the niche, the audience and the goal. Each asset ships with a hook, a body and one call to action, and with the assumptions and the missing figures written out rather than quietly filled.

Intake

Name these before writing a single word:

  • Platform. Which platform, or which set when the job is a chain or a calendar. One asset per platform, never one asset pasted everywhere.
  • Niche. What the account is about and what the user has actually done or knows first hand.
  • Audience. Who reads it, in their situation. "Founders" is a label; "solo founders who have shipped twice and stalled on their third" is an audience.
  • Goal. The one action the reader should take next: follow, reply, subscribe, finish the video, book a call.
  • Existing material. Anything to draw on: a source piece, past posts that worked, brand or voice notes, product facts, links, an audience description the user already uses.

If inputs are missing, ask once, in one message, with at most three questions. Ask for platform and goal first, then audience, then material. If the user says go, proceed and print the assumptions you used at the top of the deliverable. Never ask the user for a number you could leave as a slot, and never invent one to save the asking.

Modes

Each mode is one deliverable. Read references/formats.md for the structure of the single-asset modes and references/calendar-repurposing.md for the calendar and the chains.

Mode The ask that maps to it What you deliver
X thread "Write me a thread about X" Hook post, a stack of one-idea posts that each stand alone, a recap, a closing call to action
LinkedIn post "A LinkedIn post that brings leads" Line-one hook, short-line body that earns the click on "see more", one call to action, plus a comment variant to test
YouTube script "A YouTube script people actually finish" Spoken script: cold open, section beats with re-hooks, payoff before the ask, end-screen call to action, timestamps and B-roll notes
Short-form script "A 30 second TikTok / Reel / Short" Two-column audio and visual script, hook inside the first line, one idea per shot, loop or payoff ending
Instagram caption "A caption for this image or reel" Opening-line options, caption body, one call to action, hashtag and link-in-bio slots
Newsletter issue "Write this week's issue" Subject options, preview text, opening, body sections, one call to action, postscript
Blog article "A long-form article on X" Answer-first introduction, H2 skeleton built from the reader's questions, body with evidence slots, internal-link placeholder list, opt-in block
Podcast outline "An outline for episode X" Cold open, timed segment map, question bank, written transitions, ad-read and outro slots
30-day calendar "Plan next month of content" Day-by-day table across the named platforms, one asset per row, each with hook type, pillar and call to action
Repurposing chain "Turn this into ten posts" A source piece split into per-platform derivatives, each restructured for its platform, with a chain ledger

A request that spans modes (a calendar and the first four assets in it) is delivered as the calendar first, then the named assets.

Output format

Every single asset uses the same header and tail so a chain stays auditable:

PLATFORM · ASSET · MODE
Audience: <who>   Goal: <one action>   Assumptions: <what you guessed>

HOOK   the line or the first seconds
BODY   the asset itself, written out, not described
CTA    one action plus what happens next for the reader
TEST   what to compare once it is live
GAPS   figures or facts this skill did not supply

Worked example, shortened. Brief: freelance book editor, audience is first-time novelists, goal is newsletter sign-ups, no figures supplied.

X · first-draft checks · X thread
Audience: first-time novelists   Goal: newsletter sign-up   Assumptions: they are drafting, not querying

HOOK   Three things a first draft cannot be asked to do, and what to check instead.
BODY   1/ It cannot be judged for pacing. Check: where you skimmed. Skimming marks the
       parts you already know are soft.
       2/ It cannot tell you whether the promise landed...
       3/ It cannot show you a scene you cut and forgot.
CTA    The checklist I run on a draft, sent as a one-page sheet. Sign up: [link — supply this]
TEST   Compare replies and profile clicks on post 1 against the same week's plain
       announcement. Bookmark rate is the signal to watch: [baseline — supply this]
GAPS   Sign-up link, a line from a real edit, any count of past projects

Quality bar

Honesty:

  • No invented metrics, follower counts, revenue, testimonials, case studies, client names or competitor facts. A figure the user has not supplied becomes [metric — supply this] and appears under Gaps.
  • A claim about how a platform works is phrased as "commonly reported" or "test this", never as fact about Google, Meta, LinkedIn, TikTok or X internals. Ranking behaviour changes and this skill does not have the source.
  • Build the CTA from the reader's next action and say what happens when they take it. Do not promise it lifts reach, bookmarks or search placement; write the test that would tell the user whether it did.
  • Anything the user has not supplied about their niche that the asset depends on is fetched in the session and cited with a date, or marked [unverified].
  • A hook may only promise what the body actually delivers. Curiosity that the piece never pays off is a defect, not a technique.

Craft:

  • One idea per asset, one paragraph per beat, one call to action per asset.
  • Write for the first screen: the line a reader sees before any "see more", "read more" or the first three seconds of a video carries the whole job.
  • Platform-native means different structure, not different length. A thread reflowed into a LinkedIn post is still a thread.
  • Keep the voice the user's own. Draw words from the supplied material; do not import marketing adjectives, and no hype about the content's effect.
  • Retention is engineered with structure: open loops, re-hooks at each section, the payoff before the ask. Name the mechanism, do not claim the number.
  • The blog article is yours to structure. Internal links ship as a placeholder list of anchor text and target slots. Cluster choice, keyword assignment and cannibalization belong to keyword-guard; do not re-teach topic-cluster theory in an article.

Before handing anything over, check your own output:

  • [ ] No number appears that the user supplied or a dated source gave.
  • [ ] No "viral", "guaranteed", "will rank", "performs above average" or any other promise.
  • [ ] Every platform claim reads as "commonly reported" or as a test.
  • [ ] One CTA per asset, with what happens next for the reader.
  • [ ] Assumptions printed at the top, gaps listed at the bottom.
  • [ ] Asked at most once, in one message.
  • [ ] In a chain or calendar, no two assets share the same structure and only differ in length.

Handoffs

Signal in the request Sibling What you send
The asset sells a product, a launch, an ad, a landing page or a cart email sales-copy The brief, the audience, the angle you used, and the exact claim set so both versions say the same thing
"Fix what I already wrote", shorten, polish, match my voice wordsmith The draft, what must be preserved (facts, voice, length), and the one change wanted
Which page should rank for this query, cluster the blog, two posts compete keyword-guard Article or page list, the query each targets, the internal-link placeholder list
Should this channel exist at all, pricing, market entry, a roadmap strategy-memo The decision, the constraint, what the content plan assumed
What is actually true about a competitor, a market, a claim research-brief The subject, the question, the depth needed, the decision it informs
Reach dropped, which posts worked, my own export or numbers data-readout The export, the question, the assets it covers, the decision it informs
Write the code that posts, pulls or schedules this production-code The platform, the action, the trigger, where the data lives
A resume, LinkedIn profile rewrite or promotion narrative career-track The line the user asked for and why it is a career document, not audience content

Send the block in this shape:

HANDOFF — platform-content -> <target skill>
Goal: <one line>
Brief: <platform/audience/asset, or subject/dataset/role>
Evidence: <what the user supplied>
Assumptions: <what you assumed to continue>
Gaps: <unverified figures, missing inputs>
Next: <one line offering the target skill>

If the sibling is not installed, output the block and say which skill normally owns it.

04 同仓库更多 Skills

arm-pz
arm-pz/caffeinity-skills

tune-and-ship

Takes one published skill at a time from as-is to tuned, fact-checked, released and live, then moves to the next one in the catalog. Each run tunes the skill with tune-skill, checks every version number, flag, price and command it states against the primary source, re-reads its siblings' real SKILL.md files, polishes the listing with findable-polish, writes the version and release notes, opens one PR, and after the merge confirms the site page and marketplace copy are live. Use whenever the user wants to fine-tune their other skills, work through the catalog, give a published skill the full treatment, fact-check a skill's claims, or ship a tune instead of drafting it, even if they never say "ship". Not for a tune with no release (use tune-skill), drafts with nothing pushed (use tune-and-polish), a new skill (use skill-launch), regressions from the nightly health check (use caffeinity-routing's Descale brew), or a whole collection (use skill-estate).

★ 0
arm-pz
arm-pz/caffeinity-skills

skill-estate

Takes a whole collection of agent skill repos from scattered to published, consistent and self-maintaining, in gated phases: inventory every repo, reconcile site, marketplace and README listings, test and tune the family, close workflow gaps, wire publishing so new skill repos go live on their own, bundle a plugin with agents, add trigger tests and directory listings, and leave a maintenance loop and a map. Each phase hands off to the skill that owns it and ends in PRs. Use when someone's skills site and GitHub disagree, wants all their skill repos audited or cleaned up, wants new skills to publish themselves, or wants that cleanup repeated on another org, even if they never say estate. Not for one skill (use tune-skill), one listing (use findable), the site pipeline alone (use publish-pipeline), or weekly upkeep (use caffeinity-routing's Descale brew).

★ 0
arm-pz
arm-pz/caffeinity-skills

paper-trail

Drafts the policy pages a small product needs before it takes money or data (privacy policy, terms of service, refund policy, cookie notice) from a fact sheet about what the product actually collects, stores, shares and charges, then checks the drafts for placeholders, missing sections and facts nobody supplied. Use when someone needs a privacy policy, terms, refund or cookie page for a site, app, SaaS, newsletter or digital product, asks what legal pages they need before launch, or wants existing policy pages checked against what the product really does, even if they never say "legal". Not legal advice and not for contracts, NDAs or employment documents (use a lawyer), and not for the launch plan itself (use solo-income-blueprint) or the site build (use site-prompt).

★ 0
arm-pz
arm-pz/caffeinity-skills

course-builder

Designs a course for a group of learners (a class, cohort, workshop series, bootcamp or internal training): measurable outcomes, a module sequence with prerequisites, activities and assessments aligned to each outcome, a time budget that fits the hours available, and a rubric, then checks the plan with a script so no outcome goes untaught or unassessed. Use when someone is teaching others and asks for a curriculum, syllabus, course outline, module plan, cohort or workshop structure, or wants an existing course checked for gaps and overload, even if they never say "curriculum". Not for one person's own learning plan (use study-system), for being taught the topic (use topic-primer), or for marketing the course (use sales-copy).

★ 0
arm-pz
arm-pz/caffeinity-skills

wordsmith

Write, draft, rewrite, edit, proofread, polish and improve text while preserving the author's intended meaning, facts and voice. Use when the user wants clearer, shorter or better-toned writing; grammar, structure or flow fixes; emails, website or marketing copy, headlines, outlines or summaries; a different reading level; an accessibility, inclusivity or accuracy review; or content prepared for publication, even if they never say "edit". Also handles /wordsmith commands such as draft, critique, audit, tighten, tone and final. Not for resumes, CVs or LinkedIn profiles (use recruiter-lens if installed), not for code or code comments, and not for building .docx or .pdf files (write the text here, then use the file skill).

★ 0
arm-pz
arm-pz/caffeinity-skills

web-preflight

Use when shipping, deploying, publishing, launching, reviewing, or pre-flighting a website or API and you want a lightweight check for accidental exposure of secrets, source or configuration files, admin and debug surfaces, unsafe redirects, missing security controls, and misleading 200 responses. Trigger phrases include "before we ship", "security check", "preflight", "audit the site", "check the production URL", "is my .env exposed", and "why does /.env return 200", including when access-log or analytics evidence is supplied. Checks only sites the user owns or is authorized to assess, with bounded GET and HEAD requests and a fixed compact path set; distinguishes genuine resources from SPA catch-all and blanket WAF responses. Not for penetration testing, exploitation, credential testing, broad port scanning, or application-code review; use a dedicated authorized security-testing or code-review workflow instead.

★ 0