What does this skill do, and when should you use it?
Social Creative Builder is for marketers and agencies producing social content across multiple platforms. It rewrites one approved idea into posts, threads, captions, Xiaohongshu notes, link posts, and carousel specifications for channels such as X, LinkedIn, Threads, Instagram, and Xiaohongshu. Each version follows the relevant dated platform norm card instead of being copied verbatim across channels, while product claims remain tied to approved ledger wording. The skill produces human-publishable packages and does not post, schedule, or automate engagement.
Reads the idea, requested platforms, destination URL, narrative, claims and channel projections, voice card, dated platform norm cards, and optional calendar context; selects platform-specific hooks such as question, contrarian, number-led, story, curiosity-gap, proof-point, or POV; produces ready-to-paste posts, threads, captions, tags, link and first-comment placement, disclosures, and alt text; in carousel mode, assigns intro, content, and outro slide roles, sets character budgets, specifies Xiaohongshu 3:4, Instagram 4:5, or LinkedIn document PDF artboards, and adds safe-zone and per-slide alt-text guidance; flags unverified claims as [needs source] and emits the required handoff summary.
- A marketing lead has an approved product idea and needs native versions for X, LinkedIn, Threads, and Xiaohongshu.
- A content team wants to turn a blog section into an X thread and a professionally framed LinkedIn post.
- A social design team needs a slide-by-slide copy specification for Xiaohongshu 3:4 and Instagram 4:5 carousels.
- A brand team needs a link-post variant with platform-specific link and first-comment placement.
- A compliance-sensitive product team needs every product claim checked against approved claims-ledger wording or visibly flagged for sourcing.
How do you install this skill?
- Before publication, a human should verify each platform's current norm card, disclosure requirements, link placement, and claims ledger; missing or stale specifications must not be guessed.
- The skill reads project memory and requests writes under memory; confirm that the host actually enforces the stated private-storage, authorized-proposal, and audit controls, and retain recoverable runtime state.
- The skill declares version 18.0.0 while SECURITY.md identifies 17.x as the supported current line; confirm the actual support line and maintenance commitment before deployment.
- Mainland-China platforms are primarily delivered as manual packages; do not assume reachable official automation or analytics interfaces.
- Shell / CLI
- Network access
- Local filesystem
Python standard-library connector scripts
The skill is located at social/craft/social-creative-builder/SKILL.md in the repository. For the full collection in Claude Code, run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills and then /plugin install aaron-marketing@aaron. Other Agent Skills-compatible hosts can use npx skills add aaron-he-zhu/aaron-marketing-skills, or the repository can be cloned with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The source documents the single-skill command for generic hosts but does not provide a separate standalone installation procedure for this skill.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/aaron-he-zhu/aaron-marketing-skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/social/craft/social-creative-builder" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Turn this idea into packages for X, LinkedIn, Threads, and Xiaohongshu. Add a link-post variant for [URL] with first-comment placement: [idea].
- Carousel mode: spec a 7-slide carousel from this outline for Xiaohongshu (3:4) and IG (4:5): [outline].
Use a prompt such as “Turn this idea into packages for X, LinkedIn, Threads, and Xiaohongshu. Add a link-post variant for [URL] with first-comment placement: [idea].” For carousel work, use “Carousel mode: spec a 7-slide carousel from this outline for Xiaohongshu (3:4) and IG (4:5): [outline].” Provide the approved idea, target platforms, and relevant narrative, claims, channel, voice, and norm-card inputs. If a norm card is missing or stale, the skill routes the issue to platform-norm-profiler instead of guessing specifications.
What are this skill's strengths and limitations?
- Supports native content packages for X, LinkedIn, Threads, Xiaohongshu, and link posts.
- Explicitly prevents verbatim cross-posting and requires platform-specific adaptation.
- Builds claims traceability, source flags, disclosures, alt text, and variant labels into the deliverable.
- Includes carousel slide specifications with platform-specific artboard ratios.
- Keeps publishing human-controlled and excludes posting, scheduling, DM, and engagement automation.
- Depends on upstream narrative, claims, channel, voice, and dated norm-card inputs; missing inputs can block publish-ready output.
- Unverified product claims remain marked [needs source] and prevent publish-ready status.
- Carousel mode specifies layouts and copy but does not render images.
- Chinese platforms are handled as manual-package access classes rather than automated channels.
- The supplied source provides no standalone test suite or evidence of actual performance on each platform.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Social Creative Builder this page | 59 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
| Narrative Enablement Kit | 55 · Use with care | ★ 3k | 12d ago | Apache-2.0 |
| Platform Norm Profiler | 60 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
| Channel Portfolio Planner | 57 · Use with care | ★ 3k | 12d ago | Apache-2.0 |
| Blog Repurpose | 50 · Use with care | ★ 2.4k | 13d ago | MIT |
Unlike social-calendar-builder, it builds individual content packages rather than planning cadence. Unlike short-video-scripter, it creates posts, threads, notes, and carousel specifications rather than video beat sheets. Unlike brief-generator and content-amplifier, it does not create creator deliverable briefs or boost existing assets with paid spend. social-quality-auditor is the named downstream pre-publish review skill.
How did FollowSkills review this skill?
The skill clearly limits itself to content-package generation and forbids automatic posting, scheduling, or engagement; it requires human publication, flags unverified claims, uses authorized proposal events, and treats supplied content as untrusted data. Points are deducted because concrete permission boundaries, rollback behavior, dependency availability, and host isolation are not fully verifiable from this file, while publication confirmation remains mainly a human-control requirement.
Triggers, inputs, outputs, completion criteria, termination rules, and missing-data handoffs are reasonably explicit, with publish blocking and dependency status for failures. Static calibration requires a deduction: nothing was executed, referenced norm cards, registries, runtimes, and shared protocols were not supplied for reproduction, and key write/proposal paths cannot be reproduced; the score therefore remains below the static ceiling of 10.
Use cases, supported platforms, carousel mode, argument hint, and non-fit boundaries are clearly stated, including manual Chinese Xiaohongshu delivery. Points are deducted because platform behavior depends on externally maintained dated norm cards, input-quality boundaries and mainland-China reachability are not sufficiently established, and Chinese platforms require manual operation.
Frontmatter provides name, version, compatibility, license, author, and homepage; the document includes Quick Start, Skill Contract, Data Sources, Instructions, save rules, and next-skill guidance. Points are deducted for unclear maintenance ownership, update path, changelog, and troubleshooting; the supplied SECURITY.md says 17.x is the supported current line while the skill declares 18.0.0, creating governance inconsistency.
The claimed output is concrete: platform-ready copy, tags, disclosures, first-comment link placement, accessibility text, variant labels, and slide-by-slide carousel specifications, with an explicit prohibition on verbatim cross-posting. Static calibration caps this at 7; no representative generated output, third-party acceptance, or execution evidence was supplied to establish correctness, completeness, or marginal benefit over manual work.
The skill requires named-source citations, claims-ledger traceability, dated norm cards, and dependency offsets; the repository also supplies general CI, architecture, and audit-validator tests. Points are deducted because those tests do not cover this skill's core content-generation paths, and no independent reproduction record, third-party execution evidence, or cross-source corroboration was provided; static review caps verifiability at 5.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →