skill-builder_skill

This skill converts extracted documentation into upload-ready Claude skill packages, streamlining structure, references, and validation for rapid deployment.
  • TypeScript

92

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill jmagly/aiwg --skill skill-builder

  • SKILL.md6.7 KB

Overview

This skill transforms extracted documentation into a ready-to-upload Claude skill package with a consistent directory layout and navigable reference files. It validates input quality, categorizes content into logical sections, and produces a complete skill manifest and supporting reference documents. The skill enforces grounding and escalation checks to avoid guessing and preserves checkpoints for recovery.

How this skill works

It first validates the input directory, content formats, and output permissions, then maps pages to predefined categories (getting started, API reference, guides, concepts, advanced). It generates the main skill document, structured reference files, optional scripts and assets, and runs output validation checks. On errors it preserves state, presents diagnostics, and escalates for user guidance rather than making assumptions.

When to use it

  • After running a doc scraper or PDF extractor that produced parsed pages
  • When you need a standardized Claude skill package for upload or review
  • When documentation needs categorization into getting-started, API, and guide sections
  • Before sharing documentation with other agents or publishing to an agent directory
  • When you need recoverable, repeatable builds with checkpointing

Best practices

  • Verify the input directory and page formats before running a build
  • Provide explicit guidance if multiple formats or unclear category mappings exist
  • Set sensible options (max examples, min category pages) to avoid over/under-generating content
  • Use the grounding checklist—don’t allow automated placeholder content
  • Run a quality check after build and review flagged issues before upload

Example use cases

  • Convert a scraped project docs folder into a structured skill package for Claude
  • Turn exported API docs into an API-focused skill with indexed reference pages
  • Build a tutorial-oriented skill from a set of how-to guides and examples
  • Recover and re-run a partial build using saved checkpoints after a parsing failure
  • Generate both minimal and full templates depending on audience and publishing needs

FAQ

JSON pages and Markdown are supported by default; provide guidance if you have multiple formats or uncommon types.

What happens if categorization fails?

The build pauses, saves state, and requests a manual category mapping; do not allow automatic guesses for unclear content.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
skill-builder skill by jmagly/aiwg | VeilStrat