Repository inventory

vincentchan/ai-content-engine

Skills indexed from this repository, with install-style signals scoped to the repo.
4 skills96 GitHub stars0 weekly installsGitHubOwner profile

Overview

This skill generates new content drafts by analyzing up to five reference pieces and running a structured pipeline that deconstructs, codifies, and re-creates content variations. It automates research, context gathering, meta-prompt creation, and produces three draft variations per content type with pre-flight checklists and source tracking. Outputs are saved with timestamped filenames for traceability.

How this skill works

You provide up to five reference URLs or paste content. The skill fetches and deconstructs those sources, builds a content anatomy guide, generates context questions, and crafts a meta-prompt that runs in two phases: an interview-style context gathering (up to 10 questions) and a content-writing phase that produces three variations per requested content type. All intermediate artifacts and final drafts are saved using a consistent timestamped naming convention.

When to use it

  • You need quickly reproducible content inspired by strong examples (articles, threads, landing pages).
  • You want structured drafts that follow a proven content anatomy and psychological playbook.
  • You need multiple variations for A/B testing, repurposing, or social distribution.
  • You are preparing briefs for writers or want reproducible prompts for content teams.
  • You want traceable outputs and a clear audit trail of research and decisions.

Best practices

  • Provide 2–5 high-quality reference URLs that exemplify the tone, structure, or performance you want to emulate.
  • Answer Phase 1 context questions fully and specifically—clear goals, audience, and examples speed up quality drafts.
  • Use the saved anatomy and context files to adjust voice, hooks, and pacing before generating final drafts.
  • Keep one consistent timestamped run per project to maintain traceability across files.
  • Validate Twitter/X links are transformed for FxTwitter when sharing tweet URLs to avoid fetch failures.

Example use cases

  • Create three headline-and-intro variations for a long-form article based on top-ranking competitors.
  • Generate multiple social thread drafts and hook variants from a popular thought-leader thread.
  • Repurpose a webinar transcript into three email sequence variations with different psychological hooks.
  • Produce three landing page copy variations that follow a tested anatomy for conversion optimization.
  • Draft three newsletter versions targeting different audience segments using the same source research.

FAQ

Up to five; at least one is required to start the pipeline.

What happens if a URL fails to fetch?

The system logs failures, continues with successful fetches, and reports the failed URLs and reasons.

How many variations are produced?

Exactly three variations per requested content type are generated in the content-writing phase.

Where are outputs saved and how are they named?

All artifacts are saved in designated folders with a single run timestamp in the format YYYY-MM-DD-HHmmss for traceability.

4 skills

More from this maintainer
Other repositories and skills published under the same GitHub owner.
Skills library
Jump back to the full directory or explore grouped topics.
Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational