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 Delivery

ASecurity

Use when packaging or delivering domain skills over MCP per SEP-2640: check that a skill package carries a conformant SKILL.md (kebab-case name, description), build the skill URI (skill:// namespace and path), and verify the MCP server exposes the delivery model (resources/read and directory listing behind the directoryRead capability). SEP-2640 serves skills as resources and stays an adapter over the agentskills.io content format, an emerging spec not yet stable. Trigger: SEP-2640, skills ov...

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

Works with

claude codecursormcp

Security Analysis

A100/100

Scanned 9/27/2026

Install to Claude Code

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

Installs into .claude/skills of the current project.

Are you the author of Skill Delivery?

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

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

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

Download with Pro
Files
SKILL.md
---
name: skill-delivery
description: "Use when packaging or delivering domain skills over MCP per SEP-2640: check that a skill package carries a conformant SKILL.md (kebab-case name, description), build the skill URI (skill:// namespace and path), and verify the MCP server exposes the delivery model (resources/read and directory listing behind the directoryRead capability). SEP-2640 serves skills as resources and stays an adapter over the agentskills.io content format, an emerging spec not yet stable. Trigger: SEP-2640, skills over MCP, skill delivery, skill URI, MCP resources, directory read, skill packaging, 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)"
mcp_allowed:
  - aero-agent-skills:read
  - aero-agent-roles:read
mcp_blocked:
  - worldintel
metadata:
  domain: cross-cutting
  subdomain: sep2640
  tags: [sep-2640, skills-over-mcp, mcp, skill-delivery, resources, skill-uri]
  version: 0.1.0
  author: Aero Agent Skills
---

# SEP-2640 Skill Delivery (cross-cutting/sep2640/skill-delivery)

Use when the task is delivering skills over MCP per SEP-2640:
package conformance, skill URIs, and server readiness.

## Domain quick reference

- SEP-2640 (Skills-over-MCP, MCP working group) serves skills as
  resources: skill:// URIs, resources/read, and directory listing
  behind the directoryRead capability.
- The agentskills.io SKILL.md format remains the canonical content
  form; SEP-2640 is an adapter layer for discovery and delivery.
- A deliverable package carries a conformant SKILL.md at its root:
  kebab-case name and a description.
- Status: the SEP-2640 spec draft is not yet stable, but the Aero
  Agent Skills delivery server (packages/aero-agent-skills/lib/mcp.js)
  implements the resources model TODAY over skill:// URIs — reference
  files under references/ are served too. Same for Aero Agent Roles
  (role:// URIs), which ships as the separate `aero-agent-roles`
  package rather than from this repository.

## Workflow

1. Check the skill package for conformance (SKILL.md present,
   kebab-case name, description).
2. Build the skill URI from the namespace and skill path.
3. Verify the MCP server exposes the delivery model: skill URIs,
   resources/read, and directory listing. The in-repo servers speak
   newline-delimited JSON-RPC 2.0 over stdio (zero deps):
     node packages/aero-agent-skills/bin/aero-agent-skills.js mcp
     npx aero-agent-roles mcp    # separate package, not in this repo
   Register them with any MCP host (Hermes mcp_servers config,
   `claude mcp add`, VS Code/Cursor/Windsurf .mcp.json) and call
   resources/list + resources/read.
4. To read a skill's deep files, read the skill body first — the
   server appends a `skill files:` footer listing references/,
   scripts/, assets/ — then fetch a file at
   skill://<family>/<pack>/<leaf>/references/<file>.
5. Pin the SEP-2640 revision and note the emerging status when you
   build against the external spec; the in-repo server implements the
   resources shape so it also works as a reference implementation.

## Pitfalls

- Shipping a package without a root SKILL.md.
- A server without directoryRead claiming directory discovery.
- Treating SEP-2640 as stable while the spec is still a draft.
- Using the MCP layer as the source of truth instead of the
  SKILL.md files.
- Local-path shadowing: inside readResource, never name the local
  URI variable `path` (it shadows node:path; use relPath).
- Reference reads must stay under the skill dir — reject `..`.

## Behavior contract (gate 3)

The package-conformance, URI, and readiness logic is exercised by the
gate 3 contract test: scripts/test_delivery.py against
scripts/delivery_logic.py (stdlib unittest, offline). Run:
python3 scripts/test_delivery.py

## Compliance

- SEP-2640 is an open specification; quote with citation and note the
  status (emerging, not yet stable) per standards-map.yaml.
- compliance: STANDARDS-REF, gated: false.

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 →