Charlie Hills Social Media Skills - Setup Guide¶

Source: skills.sh (17 skills, ~17,200 combined installs) GitHub: charlie947/social-media-skills (3,357⭐, 789 forks, MIT, active Aug 2026) Category: Marketing / Social Media First Seen: Sep 9, 2026 sweep (cluster surfaced via failed-query compensation; not previously evaluated) Quality Tier: 🟡 Production with per-skill caveats (15 of 17 skills Pass all audits; voice-builder carries a Snyk CRITICAL flag, reels-scripting carries Trust Hub + Snyk Warns - see Security)

The complete skill system behind Charlie Hills' content operation - 415K+ followers across LinkedIn, Instagram, Substack, X, and YouTube with 100M+ views per year. Every skill reads shared context: voice-builder produces about-me.md and voice.md, and every other skill checks those files before drafting a line, so everything ships in one consistent brand voice.

Installation¶

npx skills add charlie947/social-media-skills

For Hermes Agent, copy into the skills directory from a clone:

git clone https://github.com/charlie947/social-media-skills.git
cp -r social-media-skills/skills/* ~/.hermes/skills/

Order matters: run voice-builder first. Every other skill reads about-me.md + voice.md and degrades without them.

Prerequisites¶

Variable	Needed for
`APIFY_API_TOKEN`	`post-scorer`, `reels-scripting`
`GOOGLE_AI_API_KEY`	`reels-scripting` (Gemini 2.5 Flash video analysis)

export APIFY_API_TOKEN=your_token
export GOOGLE_AI_API_KEY=your_key

The image skills (gemini-infographic, gemini-carousel, quote-post, youtube-thumbnail, profile-optimizer) output ready-to-paste Gemini prompts - no API key needed.

Skill Roster¶

Group	Skills
Voice foundation	`voice-builder` (interview + writing samples → `about-me.md`, `voice.md`) · `newsletter-voice` (newsletter rules on top)
LinkedIn	`profile-optimizer` (full profile rebuild) · `post-writer` (drafts in your voice) · `graphic-designer` (HTML/CSS graphic or AI infographic) · `post-formatter` (PAS/AIDA/BAB/STAR/SLAY frameworks) · `hook-generator` (6 hook variations per topic) · `post-scorer` (scores drafts against post history) · `content-matrix` (pillars × formats ideation) · `niche-research` (7-day niche research) · `gemini-infographic` · `gemini-carousel` · `quote-post` (two-step quote workflow)
Instagram Reels	`reels-scripting` (Apify + Gemini 2.5 Flash reference analysis, newsletter-aligned script)
YouTube	`youtube-thumbnail` (title → Gemini thumbnail prompt)
Community	`pinned-comment` (meme-style pin + image prompt)
Analytics	`analytics-dashboard` (LinkedIn export → dashboard + 5 recommendations)

Quick Start¶

Install skills, set env vars for the analytics/reels skills
Run voice-builder: answer a short interview, paste 3-5 writing samples
Draft: "Write me a post about AI agents" → post-writer; "Score this draft against my history" → post-scorer; "Make me a carousel from this" → gemini-carousel

CorpusIQ Use Cases¶

Use Case	How
Brand voice lock	CorpusIQ's content voice rules get a working voice-profile mechanism (`about-me.md` + `voice.md`) that every post draft reads - consistency across all platforms
LinkedIn growth ops	Operator-facing LinkedIn posts with hook variants and history-scored drafts - the lead-generation channel
Newsletter-first repurposing	`newsletter-voice` + per-platform skills mirror the CorpusIQ belief-bridge content flow (one source piece, platform-adapted output)
Reels/Shorts pipeline	`reels-scripting` feeds the UGC video pipeline script selection; `youtube-thumbnail` feeds thumbnail prompts

Integration with CorpusIQ¶

CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.

This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.

See what CorpusIQ connects

Limitations / Verification¶

Voice profile files live in the project - about-me.md/voice.md are written into the working project, not the agent install; keep them under version control to reuse across sessions
External deps for 2 skills - post-scorer and reels-scripting need Apify; reels-scripting also needs a Gemini key
Security flags on 2 skills - see below; install the other 15 freely

# Verify skill installed
ls ~/.hermes/skills/voice-builder/SKILL.md ~/.hermes/skills/post-writer/SKILL.md

Security¶

Spot-checked Sep 9, 2026: post-writer, hook-generator Pass all three audits. Two flagged skills:

Skill	Trust Hub	Socket	Snyk	Note
`voice-builder`	Pass	Pass	Fail (CRITICAL)	E004: contradictory instruction detected - the skill states it must produce `voice.md` and later says "Do not produce an voice.md file"; Snyk reads it as a hidden directive (0.80). Also W011 (third-party content ingestion - inherent to its interview workflow). Use with care; review SKILL.md before running, or build the voice files manually
`reels-scripting`	Warn	Pass	Warn	W011 third-party content exposure via Apify/Gemini reference analysis - inherent to the workflow

The remaining 15 skills carry clean audits. The publisher is a well-known creator (3.4K⭐ repo, public newsletter, contributions welcome) - the CRITICAL flag reads as a wording bug, not injected content, but treat the skill with the usual untrusted-input discipline and review its instructions before first run.

← Skills Catalog | Marketplace → Powered by CorpusIQ