Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Skill Authoring

ASecurity

Use when you must author a NEW conformant SKILL.md for the agentskills.io format under SEP-2640-style delivery: build the frontmatter template from the required fields (name kebab-case matching the folder, description with action and use-when and trigger clauses, Apache-2.0 license, compliance flag, standards references, gated boolean, metadata version and author), then run the deterministic pre-publish conformance check that reports missing or invalid required fields before a leaf is publish...

2 stars
0 votes
0 copies
0 views
Added 9/27/2026
ai-agentspythongit

Works with

claude codemcp

Security Analysis

A100/100

Scanned 9/27/2026

Install to Claude Code

$npx -y skills add ashfordeOU/aero-agent-skills --skill skill-authoring --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Skill Authoring?

Add the live security badge to your README — it updates automatically with every re-scan.

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

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: skill-authoring
description: "Use when you must author a NEW conformant SKILL.md for the agentskills.io format under SEP-2640-style delivery: build the frontmatter template from the required fields (name kebab-case matching the folder, description with action and use-when and trigger clauses, Apache-2.0 license, compliance flag, standards references, gated boolean, metadata version and author), then run the deterministic pre-publish conformance check that reports missing or invalid required fields before a leaf is published. SEP-2640 stays an emerging spec, so the authoring discipline targets the stable agentskills.io content surface. Trigger: skill authoring, SKILL.md template, frontmatter, kebab-case name, pre-publish conformance check, SEP-2640, agentskills.io."
license: Apache-2.0
compliance: STANDARDS-REF
standards:
  - id: sep-2640
    reference-only: true
gated: false
domain: cross-cutting
pack: cross-cutting
compatibility: "agentskills.io SKILL.md; any SKILL.md host (Claude Code, Hermes, OpenClaw)"
metadata:
  domain: cross-cutting
  subdomain: sep2640
  tags: [sep-2640, skill-authoring, skill-md-template, frontmatter, kebab-case, conformance-check, pre-publish]
  version: 0.1.0
  author: Aero Agent Skills
---

# SEP-2640 Skill Authoring (cross-cutting/sep2640/skill-authoring)

Use when the task is authoring a new conformant SKILL.md: building the
frontmatter template, checking the kebab-case name rule, writing the
router description, and running the pre-publish conformance check.

## Domain quick reference

- SEP-2640 delivery is skills over MCP in the agentskills.io content
  form; the spec is emerging, so authoring targets the stable
  conformance surface: required frontmatter fields, name shape,
  description discipline, license, compliance, standards, gated, and
  metadata.
- The required top-level fields are name, description, license,
  compliance, standards, gated, and metadata. A candidate missing any
  of them fails the pre-publish check.
- The name must be kebab-case (lowercase letters, digits, single
  hyphens) and equal the leaf folder name; the description must carry
  an action clause, a 'Use when' clause, and a 'Trigger' keyword with
  at least two trigger keywords, within the 50-150 word and 1024 char
  budgets.
- The license must equal Apache-2.0; compliance must be one of none,
  ITAR-GATED, EAR-GATED, STANDARDS-REF; standards must be non-empty;
  gated must be a boolean; metadata must carry version and author.
- The pre-publish conformance check is deterministic, offline, and
  stdlib-only: same candidate, same findings.

## Workflow

1. Draft the frontmatter from the template: name (kebab-case, matching
   the folder), description (action + use-when + trigger), license,
   compliance, standards, gated, metadata.
2. Run the conformance check on the candidate SKILL.md text.
3. Read the findings list: every problem is a missing or invalid
   required field (name shape, folder match, description budgets and
   clauses, license, compliance, standards, gated, metadata).
4. Fix the candidate until the findings list is empty.
5. Publish only a candidate with zero findings.

## Pitfalls

- A name that is not kebab-case (uppercase, spaces, double hyphens).
- A name that does not match the leaf folder (the router resolves
  skills by folder path).
- A description without a 'Use when' clause or fewer than two trigger
  keywords (the description is the router; a weak one routes poorly).
- Claiming conformance while a required field is missing; the check
  is boolean per field and one failure fails the candidate.

## Behavior contract (gate 3)

The authoring logic is exercised by the gate 3 contract test:
scripts/test_authoring.py against scripts/authoring_logic.py (stdlib
unittest, offline). Run:

    python3 scripts/test_authoring.py

Attribution

ashfordeOUashfordeOU
View sourceMore from ashfordeOU →
SSkills DirectorySkills Directory

Know which skills are safe — weekly.

Best new skills + every skill we flagged as malicious. From the team that scanned 103,619.

Join free

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 (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Know which skills are safe — weekly.

Best new skills + every skill we flagged as malicious. From the team that scanned 103,619.

Join free

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1074701 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

694821 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

691 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →