What does this skill do, and when should you use it?
This skill lives at .github/skills/agnostic-prompt-standard/_template/SKILL.md in jsburckhardt/devcontainer-features and is essentially the official template skeleton of the agnostic-prompt-standard framework (spec_version 1.0, framework_revision 1.0.0). It performs no concrete task itself: the body is a placeholder one-paragraph overview plus a fixed directory-layout convention (references, assets, processes, guides, scripts). Its value is giving authors a copy-ready starting point that keeps new skills consistent with the APS spec's file structure and metadata. Whether to adopt it depends on whether you plan to write your own skill under the APS standard.
Provides SKILL.md frontmatter with APS metadata fields (spec_version, framework_revision, last_updated); defines a standard skill directory layout: references/ (normative spec docs), assets/constants and assets/formats (constants and format contracts), processes/ (executable APS process docs), guides/ (prose docs for humans and agents), scripts/ (optional build/compile/lint scripts); includes a References section template pointing to references/00-<name>.md as the first normative document; contains only placeholders (<Skill Name>, <One-sentence skill description>, etc.) — no actual instructions or scripts.
- Skill authors: starting a new skill aligned with the agnostic-prompt-standard framework (v1.1.16 release line) by copying and filling in this skeleton
- Team standardization: enforcing uniform metadata fields and directory structure across a repository's skills to reduce review friction
- Documentation reorganizers: using the ready-made references/assets/processes convention to turn scattered prompt material into a structured skill
- End users expecting a working capability out of the box — the body is entirely placeholders and executes nothing
- Authors not adopting the APS spec — if you only need a minimal name + description SKILL.md, this directory convention is overhead
How do you install this skill?
- The assessed file is an unfilled template (placeholder name/description/author) and cannot be used as a runnable skill; do not invoke or trigger based on it.
- The referenced references/00-*.md, processes/, and scripts/ contents are not provided in evidence; actual capability and safety are unknown.
- The publisher is unverified by the FollowSkills registry, and the repo MIT license is attributed to Microsoft Corporation — provenance should be independently confirmed.
- Core content is hosted on GitHub with no Chinese-language support or mainland-China reachability guarantees.
- Local filesystem
The source documents no install command for the skill itself (the repo README only covers installing its devcontainer features collection, which is unrelated to this skill). The practical route is copying the file from GitHub into your skills directory. Anything undocumented (marketplace or scripted install) should not be assumed.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/jsburckhardt/devcontainer-features.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/.github/skills/agnostic-prompt-standard/_template" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Draft a new SKILL.md skeleton following the directory layout in .github/skills/agnostic-prompt-standard/_template/SKILL.md
- Reorganize my prompt docs to the APS template: specs in references/, constants in assets/constants/, processes in processes/
- Check whether my SKILL.md frontmatter matches the spec_version 1.0 field requirements
There is no trigger flow or executable step: this is a static template. Usage consists of replacing the placeholders (name, description, author, last_updated) with your content and populating the declared subdirectories (references/, assets/, processes/, guides/, scripts/). No commands, options, or slash commands are shown in the source.
What are this skill's strengths and limitations?
- Conforms to the open Agent Skills standard: plain YAML frontmatter with name + description, no platform-specific mechanisms, highly portable
- Ships a clear directory-layout convention (references/assets/processes/guides/scripts) that helps teams keep skills consistent
- MIT licensed — free to copy and modify
- Pure template with no functionality or example content; its value hinges entirely on whether you adopt the APS spec
- All metadata are unfilled placeholders — no author name or last_updated date, so maintenance status is unverifiable
- The repository's focus is devcontainer features; this template rides along, and the surrounding README does not document it
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| APS Skill Template (agnostic-prompt-standard skeleton) this page | 28 · Not recommended | ★ 38 | 18d ago | MIT |
| Template Skill (Mini Agent template-skill) | Insufficient evidence | ★ 3.1k | 8mo ago | MIT |
| Skill Creator | 61 · Recommended | ★ 20k | 14d ago | MIT |
| Effect JSDoc Authoring Skill | 77 · Strongly recommended | ★ 28k | 11d ago | MIT |
| QuantMind Contributor Workflow | 61 · Recommended | ★ 3.1k | 2mo ago | MIT |
How did FollowSkills review this skill?
This path is an APS template containing only placeholders (<skill-id>, <Your Name>, placeholder date); it has no scripts, external effects, or permission claims and thus no red-line risk, but authorship, data-flow, and source attribution are all empty placeholders — deducted to below-mid range.
The file contains no executable instructions or claimed functionality; no tests, edge cases, or failure feedback are assessable, only a declared directory layout, and the referenced references/00-*.md is not visible in evidence — deducted to the low band.
No scenarios, inputs/outputs, triggers, or non-fit boundaries; the description is a placeholder. The layout implies it targets APS spec authors, not end users — deducted to the low band.
Well-structured: complete front-matter fields (license: MIT, spec_version, framework_revision) and layered directories (references/assets/processes/guides/scripts) show progressive-disclosure intent, but all fields are unfilled with no versioning or maintenance ownership — deducted to the midpoint.
As a template it may help scaffold APS skills, but it delivers no concrete result, its value claim is unverifiable, and output is zero — deducted near the floor.
Only placeholder text and a directory listing; no traceable claims and no third-party execution evidence, so the static-read cap applies — low score.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →