v1.2.1 — rename template/SKILL.md to template/TEMPLATE.md

The skills.sh CLI globs **/SKILL.md across the repo when discovering
installable skills, so the template was appearing as a second skill
candidate ("your-skill-name") in the npx skills add picker. Renaming
the template avoids the glob while keeping the starter content
discoverable to contributors.

Also rewrote the template body to make it self-documenting — explains
why the file is named TEMPLATE.md, how to use it (copy the inner
SKILL.md block into your new skills/<name>/ folder), and links to
related docs.

- git mv template/SKILL.md template/TEMPLATE.md
- .github/workflows/lint.yml: required-files check now expects
  template/TEMPLATE.md
- README.md: file tree updated; note about naming added
- .claude-plugin/marketplace.json: v1.2.1
- skills/design-engineering/SKILL.md: frontmatter version 1.2.1
- CHANGELOG.md: v1.2.1 entry

Author: HKTITAN

.claude-plugin/marketplace.json

@@ -11,7 +11,7 @@
   "homepage": "https://github.com/HKTITAN/design-engineering",
   "repository": "https://github.com/HKTITAN/design-engineering",
   "license": "MIT",
-  "version": "1.2.0",
+  "version": "1.2.1",
   "plugins": [
     {
       "name": "design-engineering",

.github/workflows/lint.yml

@@ -65,7 +65,7 @@ jobs:
             "CODE_OF_CONDUCT.md"
             "SECURITY.md"
             "spec/agent-skills-spec.md"
-            "template/SKILL.md"
+            "template/TEMPLATE.md"
             ".claude-plugin/marketplace.json"
           )
           for f in "${required[@]}"; do

CHANGELOG.md

@@ -4,6 +4,14 @@ All notable changes to this skill are recorded here. Format follows [Keep a Chan
 
 ## [Unreleased]
 
+## [1.2.1] — 2026-05-22
+
+### Fixed
+
+- `template/SKILL.md` renamed to `template/TEMPLATE.md`. The skills.sh CLI globs `**/SKILL.md` across the repo when discovering installable skills, so the template file was appearing as a second skill candidate (`your-skill-name`) in the install picker. Renaming sidesteps the glob while keeping the starter content discoverable to contributors. ([#install-picker-fix](https://github.com/HKTITAN/design-engineering/commit/v1.2.1))
+- `.github/workflows/lint.yml` required-files check updated to reference `template/TEMPLATE.md`.
+- README file tree and `marketplace.json` version updated accordingly.
+
 ## [1.2.0] — 2026-05-21
 
 Adds two more named sources mirrored offline in `spec/` — the full [Agent Skills specification](https://github.com/agentskills/agentskills) and [Google Labs' design.md format](https://github.com/google-labs-code/design.md) — plus integrates [Agentation](https://www.agentation.com) as both an installer-recommended companion tool and a skill-graph node.
@@ -119,7 +127,8 @@ Initial public release of the `design-engineering` skill graph.
 npx skills add HKTITAN/design-engineering
 ```
 
-[Unreleased]: https://github.com/HKTITAN/design-engineering/compare/v1.2.0...HEAD
+[Unreleased]: https://github.com/HKTITAN/design-engineering/compare/v1.2.1...HEAD
+[1.2.1]: https://github.com/HKTITAN/design-engineering/compare/v1.2.0...v1.2.1
 [1.2.0]: https://github.com/HKTITAN/design-engineering/compare/v1.1.0...v1.2.0
 [1.1.0]: https://github.com/HKTITAN/design-engineering/compare/v1.0.0...v1.1.0
 [1.0.0]: https://github.com/HKTITAN/design-engineering/releases/tag/v1.0.0

README.md

@@ -140,7 +140,7 @@ design-engineering/
 ├── spec/
 │   └── agent-skills-spec.md           ← pointer to the Agent Skills specification
 ├── template/
-│   └── SKILL.md                       ← starter template for a new skill
+│   └── TEMPLATE.md                    ← starter template (not SKILL.md so the CLI ignores it)
 └── skills/design-engineering/
     ├── SKILL.md                       ← thin Map of Content (~90 lines)
     ├── evals/                         ← Perplexity Step-0 evals

skills/design-engineering/SKILL.md

@@ -4,7 +4,7 @@ description: Load when reviewing UI code, designing a component or page layout,
 license: MIT
 metadata:
   author: HKTITAN
-  version: "1.2.0"
+  version: "1.2.1"
   graph: true
 ---
 

template/SKILL.md

@@ -0,0 +1,63 @@
+# Skill template
+
+This file is a starter template for adding a new skill to this repo. It is **deliberately named `TEMPLATE.md` (not `SKILL.md`)** so the [skills.sh](https://skills.sh) CLI does not pick it up as an installable skill.
+
+## How to use this template
+
+1. Decide the new skill's name — lowercase, hyphenated (e.g. `my-new-skill`).
+2. Create the folder `skills/<my-new-skill>/`.
+3. Copy the SKILL.md template block below into `skills/<my-new-skill>/SKILL.md`.
+4. Update the frontmatter — `name` (must match folder), `description` (the "Load when…" line), `metadata.author`.
+5. Decide if you're using the skill-graph pattern. If yes → create `references/<theme>/` subfolders and atomic nodes; keep `graph: true`. If no → write all content directly in `SKILL.md`; set `graph: false`; stay under 500 lines per the [agentskills spec](../spec/agent-skills-spec.md).
+6. Add an entry to `.claude-plugin/marketplace.json` so the new skill is published as a Claude Code plugin.
+7. Add a CI required-files row in `.github/workflows/lint.yml` if you want CI to enforce its presence.
+
+## SKILL.md template
+
+Copy from the next line down through the end of this file into your new `skills/<my-new-skill>/SKILL.md`:
+
+```markdown
+---
+name: my-new-skill
+description: Load when ___. One or two sentences describing the user queries that should trigger this skill. Use real phrasing, not workflow summary. Examples — "designing X", "reviewing Y", "asking Z".
+license: MIT
+metadata:
+  author: YOUR_NAME
+  version: "0.1.0"
+  graph: false
+---
+
+# my-new-skill
+
+## When to use this skill
+
+(Optional — usually the frontmatter description is enough. Add this section only if you want explicit decision-tree guidance for the agent.)
+
+## How to use this skill
+
+(For graph-style skills, this becomes a Map of Content section with wikilinks to atomic nodes.)
+
+## Body
+
+(Replace this section with the actual skill content. Be terse. Every sentence costs tokens in every session.)
+
+## Gotchas
+
+(Append-only list of failure cases. Each entry: `[YYYY-MM-DD] one-line description → fix in [[node-name]]`.)
+
+---
+
+Built following the [Agent Skills specification](https://agentskills.io/specification).
+```
+
+## Why TEMPLATE.md not SKILL.md
+
+The `skills.sh` CLI globs `**/SKILL.md` across the repo when discovering installable skills. A file at `template/SKILL.md` would appear in the install picker as a second skill candidate alongside the real one, which is confusing for installers.
+
+Naming this file `TEMPLATE.md` keeps the same starter content available to contributors while avoiding the false-positive in the CLI picker.
+
+## Related
+
+- [`AGENTS.md`](../AGENTS.md) — repo-level conventions when editing skills.
+- [`CONTRIBUTING.md`](../CONTRIBUTING.md) — PR conventions, node format, length targets, cross-model testing.
+- [`spec/agent-skills-spec.md`](../spec/agent-skills-spec.md) — full Agent Skills specification (offline mirror).