prompts
This Claude Code skill enables automated creation of reusable skill files from reference documents by reading source material and exploring related codebase patterns. Use it when you need to distill documentation, guidelines, or process instructions into structured, actionable knowledge that AI assistants can apply to code generation and workflow tasks. The skill extracts essential information and generates properly formatted skill files with frontmatter triggers for keyword and file-based activation.
git clone --depth 1 https://github.com/aspenkit/aspens /tmp/prompts && cp -r /tmp/prompts/src/prompts/add- ~/.claude/skills/promptsadd-skill.md
Create a **skill file** from a reference document. You have Read/Glob/Grep tools for codebase context.
{{skill-format}}
## Your task
Read the reference document, then explore the codebase for related files and patterns. Synthesize a skill capturing the essential knowledge an AI assistant needs.
## Output format
Return exactly one file:
<file path="{{skillPath}}">
[full skill content]
</file>
## Rules
1. **Extract actionable knowledge.** Focus on what an AI needs to write correct code or follow correct processes — not background reading.
2. **Be specific.** Use actual file paths, commands, and patterns from the reference doc and codebase.
3. **Do NOT emit a `## Activation` section.** Trigger metadata belongs in the `triggers:` frontmatter field, not in a markdown section.
4. **Emit `triggers:` in the frontmatter** with `files:` (array of globs matching the key files for this domain) and `keywords:` (array of terms that signal this skill is relevant). Example:
```yaml
triggers:
files:
- app/deps.py
- app/api/v1/auth.py
keywords:
- auth
- jwt
- token
```
5. **Keep it concise.** 30-60 lines. Distill the reference document down to its essential rules and patterns.
6. **Use the exact output format.** One `<file>` tag with the path shown above.LLM-powered injection of project context into installed agent templates via `aspens customize agents`
>
Core conventions, tech stack, and project structure for aspens
Claude/Codex CLI execution layer — prompt loading, stream-json parsing, file output extraction, path sanitization, skill file writing, and skill rule generation
Top-level Commander wiring, welcome screen, missing-hook warning, CliError exit handling, and the public programmatic API surface
Multi-target output system — target abstraction, backend routing, content transforms for Codex CLI and future targets
Context health analysis — freshness, domain coverage, hub surfacing, drift detection, LLM-powered interpretation, and auto-repair for generated agent context
Incremental skill updater that maps git diffs to affected skills and optionally auto-syncs via a post-commit hook