Skills
Reusable workflows, knowledge, and prompt templates invoked on demand or auto-loaded when relevant
Extension Name Skills
Vendor Terms agent skills, SKILL.md, skill:// URI
Scopes
Project / Repo RootUser / Home
Interfaces IDE, CLI
Availability current (explicit)
Trust Model Text-only instruction packages that cannot execute code on their own; if agent has write/shell tools enabled, it could execute commands referenced in skills; no isolation between skills
Notes Auto-activate via progressive disclosure (name+description at startup, full content on demand). IDE supports slash command invocation; CLI activation is automatic only. Workspace takes priority over global on conflicts. Custom agents must explicitly include skills via skill:// URIs. IDE supports importing skills from GitHub repos or local folders.
Extension Name Skills (knowledge and workflows)
Vendor Terms skills, SKILL.md, reference skills, action skills, Agent Skills standard, bundled skills
Scopes
Project / Repo RootUser / HomeOrganization / Enterprise
Interfaces terminal CLI, VS Code, Desktop, web
Availability current (explicit)
Trust Model Reusable knowledge and workflows; auto-loaded when relevant or invoked via /name; bundled skills like /simplify, /batch, /debug included; enterprise deployment via managed settings
Notes Reference skills provide knowledge, action skills trigger workflows; can run in isolated context via subagents. Follows Agent Skills open standard. Rich frontmatter: model, effort, context:fork, agent, hooks, paths, allowed-tools. Dynamic context injection via shell preprocessing. Auto-discovered from nested directories (monorepo support). Skill content persists with auto-compaction budget (5K tokens each, 25K combined).
Extension Name Agent Skills
Vendor Terms agent skills, skill folders, SKILL.md, allowed-tools, open standard
Scopes
Project / Repo RootUser / Home
Interfaces GitHub.com, VS Code, CLI, Copilot SDK
Availability current (explicit)
Trust Model Loaded when relevant to specialized tasks rather than continuously; can include scripts and resources; allowed-tools frontmatter pre-approves tool usage
Notes VS Code, GitHub.com, CLI, and SDK (public preview) supported; JetBrains in preview; org/enterprise scopes coming soon; based on open Agent Skills standard (agentskills.io)
Extension Name Skills
Vendor Terms agent skills, SKILL.md, skill folders, .agents/skills, agents/openai.yaml, $skill-creator, $skill-installer
Scopes
Subdirectory / FolderProject / Repo RootUser / HomeMachine / Admin
Interfaces CLI, IDE extension, app
Availability current (explicit)
Trust Model Reusable workflows and domain expertise; skills are the authoring format, plugins are the distribution unit; allow_implicit_invocation policy controls auto-activation
Notes Progressive disclosure: metadata loaded at start, full content on selection. Six scope tiers: REPO (CWD, parent dirs, root via .agents/skills/), USER (~/.agents/skills/), ADMIN (/etc/codex/skills), SYSTEM (bundled). Two invocation modes: explicit ($-mention or /skills) and implicit (auto-matched by description). Built-in $skill-creator and $skill-installer. Builds on open Agent Skills standard (agentskills.io). Distributable via plugins.