Skip to content
Back to skills

Catalog Authoring

ASecurity

Authors catalog sources (skills, fragments, styles) in YAML and projects them. Use when adding a new skill or fragment to the catalog, or before editing a generated file.

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 5, 2026
developmentgit

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned November 5, 2026

npx -y skills add niksavis/handily --skill catalog-authoring --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Catalog Authoring?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Catalog Authoring
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/niksavis-catalog-authoring/badge)](https://www.skillsdirectory.com/skills/niksavis-catalog-authoring)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: catalog-authoring
description: Authors catalog sources (skills, fragments, styles) in YAML and projects them. Use when adding a new skill or fragment to the catalog, or before editing a generated file.
---
<!-- Generated by `basicly skills-build` from skill.yaml. Do not edit; edit the source. -->

# Catalog Authoring

Write catalog sources in YAML and project them to the files that coding agents load.

## Rules

- **Never name a source for discovery.** Agents load `SKILL.md`, `AGENTS.md`, `CLAUDE.md`, `*.instructions.md`, `*.mdc` and `GEMINI.md` by file name, so a source is `skill.yaml`, `<id>.fragment.yaml` or `style.yaml`. `basicly catalog lint` refuses a `.md` source.
- **Verify where the agent reads the guidance, not the file you edited.** A host can substitute or fall back, so the projection is not what the agent gets.
- **Apply the deletion test to each bullet.** If its removal causes no mistake, leave it out.
- **Tag a source with `technologies` only when it is useless outside that technology.** An untagged source ships everywhere.
- **Keep local sources out of `.basicly/core/` in a consumer repository.** `basicly install` overwrites `.basicly/core/`. Put a fragment in `.basicly-local/fragments/`.
- **Edit a generated file through its source.** `.claude/skills/`, `.claude/rules/*.md`, `.agents/skills/` and the always-on files are projections. Start a source with `basicly catalog new <fragment|skill> <name>`, check it with `basicly catalog lint`, then run `basicly build`. List the sources of one kind with `basicly catalog list skill`, and run `basicly catalog verify` for duplicates, contradictions and scope before a catalog commit.
- **Put a Claude-only prompt, command or runbook in `.claude/*.md`.** No catalog source projects it.
- **Add no copy under `.github/instructions/`.** Copilot then loads the same rule twice. Prefer a skill to a prompt file, and put a custom agent in `.github/agents/*.agent.md`.

## Read when needed

- Before creating or changing a source type, read [sources](references/sources.md) for its schema, placement, overlay and host projection.
- Before composing or delivering guidance, read [projection](references/projection.md) for wording, checks and effective-host readback.
- Completion means the source and delivered host files agree, with local overlays preserved; report actual consumer checks.

Files in this skill

  • SKILL.md2.3 KB
  • references/projection.md1.6 KB
  • references/sources.md2.9 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.