langchain-ai avatar

Skill Creator

This is meta tooling for building skills themselves. It walks you through the directory structure (user vs project scope, precedence rules), the anatomy of SKILL.md with its frontmatter and body, and when to use bundled resources like scripts, references, and assets. The core philosophy is progressive disclosure: keep metadata lean since it's always loaded, put essential instructions in SKILL.md (under 500 lines), and split detailed docs into reference files. Useful if you're creating skills regularly and want to understand the design patterns, though honestly most of this you'd pick up from reading a few existing skills. The emphasis on token efficiency and avoiding context bloat is the most practical takeaway.

skills.sh installs
1.3K
GitHub stars
31.2K
last updated
Aug 21, 2026

Install

$ npx -y skills add langchain-ai/deepagents --skill skill-creator --agent claude-code

Installs into .claude/skills of the current project.

Files

Skill Creator

Skill Location for Deep Agents Code

Deep Agents Code loads skills from five sources, listed here from lowest to highest precedence:

#DirectoryScopeNotes
0<package>/built_in_skills/Built-inShips with Deep Agents Code
1$DEEPAGENTS_HOME/<agent>/skills/User (deepagents alias)Default for dcode skills create
2~/.agents/skills/UserShared across agent tools
3.deepagents/skills/Project (deepagents alias)Default for dcode skills create --project
4.agents/skills/ProjectShared across agent tools

<agent> is the agent configuration name (default: agent). When two directories contain a skill with the same name, the higher-precedence version wins — project skills override user skills, and any user or project skill overrides built-in skills.

Example directory layout:

$DEEPAGENTS_HOME/agent/skills/     # user skills (lowest precedence)
├── skill-name-1/
│   └── SKILL.md
└── ...

<project-root>/.deepagents/skills/   # project skills (higher precedence)
├── skill-name-2/
│   └── SKILL.md
└── ...

Core Principles

Concise is Key

Sponsors

Details

Repository
github.com/langchain-ai/deepagents
Categories
DevOps & CI/CD, Documentation, AI & Agent Building, Skill Development, Automation & Workflows, Design & UI/UX
First seen
May 16, 2026

More from langchain-ai/deepagents

All 12
1alirezarezvani avatarsenior-securityalirezarezvani/claude-skillsUse when the user asks for STRIDE threat modeling, DREAD risk scoring, data-flow-diagram threat analysis, or a quick secret scan — or when a security request ne1.3Kskills.sh installs
2quodsoler avatarue-character-movementquodsoler/unreal-engine-skillsUse this skill when working with character movement, CharacterMovementComponent, CMC, movement modes, walking, falling, swimming, flying, custom movement, netwo1.3Kskills.sh installs
3nvidia avatarcuopt-server-commonnvidia/skillscuOpt REST server — what it does and how requests flow. Domain concepts; no deploy or client code.1.3Kskills.sh installs
4tursodatabase avatardebuggingtursodatabase/tursoHow to debug tursodb using Bytecode comparison, logging, ThreadSanitizer, deterministic simulation, and corruption analysis tools1.3Kskills.sh installs
5nexscope-ai avatarecommerce-content-marketingnexscope-ai/ecommerce-skillsE-commerce content marketing strategy planner. Generates content calendars, topic ideas, and platform-specific strategies by analyzing customer reviews, trends,1.3Kskills.sh installs
6digitalsamba avatarmoviepydigitalsamba/claude-code-video-toolkitPython video composition with moviepy 2.x — overlaying deterministic text on AI-generated video (LTX-2, SadTalker), compositing clips, single-file build.py vide1.3Kskills.sh installs