Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/laurigates/claude-plugins/plugin-authoringnpx skills add laurigates/claude-plugins --skill plugin-authoringgit clone --depth 1 https://github.com/laurigates/claude-pluginsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00059 | $0.01993 |
| Opus 5 | $0.00030 | $0.00996 |
| Sonnet 5 | $0.00012 | $0.00399 |
| Haiku 4.5 | $0.00006 | $0.00199 |
Grade A, and why
plugin-authoring scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/plugin-authoring
The authoring procedures for this marketplace: how to create a skill, how to create a plugin, what to update when either changes, and how to delete one without leaving dangling metadata.
Promoted out of CLAUDE.md (issue #2140) because all four are procedures with
a clear trigger — they do not need to be resident when the user is debugging a
hook. CLAUDE.md keeps only the repo blurb, the rules index, and the gotchas.
Detailed patterns live in the rules this skill names; it is the sequence, not a second copy of them.
Creating New Skills
See .claude/rules/skill-development.md for detailed patterns.
Note (Claude Code 2.1.157): plugins placed in
.claude/skillsare now auto-loaded without a marketplace entry — handy for local or quick one-off plugins. This repo's published plugins still use the full marketplace + release-please lifecycle described in Plugin Lifecycle below.
Quick Start
- Create skill directory:
mkdir -p <plugin>/skills/<skill-name> - Create
skill.mdwith YAML frontmatter:--- name: <Skill Name> description: <1-2 sentence description> allowed-tools: Bash, Read, Grep, Glob, TodoWrite created: YYYY-MM-DD modified: YYYY-MM-DD reviewed: YYYY-MM-DD --- - Follow content structure: Core Expertise → Commands → Patterns → Quick Reference
- Include agentic optimizations table
- Update all metadata files (see Plugin Lifecycle section)
Skill Granularity Decision
| Choose... | When... |
|---|---|
| Single skill | Operations are related and share context |
| Multiple skills | Distinct workflows, different user intents |
Example: bun-package-manager (deps) vs bun-development (run/test/build)
Creating User-Invocable Skills
Skills are invocable via /plugin:skill-name syntax. See .claude/rules/skill-naming.md for naming conventions.
- Create skill directory:
mkdir -p <plugin>/skills/<skill-name> - Create
SKILL.mdwith YAML frontmatter:--- name: <skill-name> description: What it does. Use when... args: <arg-spec> allowed-tools: Bash, Read argument-hint: human hint created: YYYY-MM-DD modified: YYYY-MM-DD reviewed: YYYY-MM-DD --- - Include: Context → Execution → Post-actions
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 186 lines · 59 tokens per session scan A 5e2800c91861
plugin-authoring is a skill published in the GitHub repository laurigates/claude-plugins (54 stars, last pushed 2d ago), licensed MIT. It adds 59 tokens to every session and 1,993 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
plugin-authoring
Best practices for writing skills, commands, and agents: descriptions, structure, token efficiency, frontmatter, tool scoping, and common authoring mistakes. Activate when creating or editing a SKILL.md, a command file, or an agent file, or when the user asks to "write a skill", "add a command", "create an agent", or…
dispatching-parallel-agents
Patterns for effective subagent delegation and parallel execution. Use when a task decomposes into independent subtasks, when research spans multiple areas, when building features that require coordinated specialist work, or when sequential execution is wasting time. Covers work decomposition patterns (fan-out/fan-in…
accessibility
Web accessibility patterns and WCAG 2.1/2.2 compliance for inclusive user interfaces. Use when the user asks to build accessible components, audit a UI for a11y issues, fix screen reader problems, implement keyboard navigation, check color contrast ratios, add ARIA attributes, create accessible forms, or establish…
worktree-ops
Manage git worktrees for isolated parallel development sessions. Create, list, switch between, and remove worktrees with safety checks for uncommitted changes and unpushed commits. Use when the user wants to work on multiple branches simultaneously, run parallel sessions, or isolate experimental changes.
composer-dependencies
Composer dependency management playbook for safe, systematic package updates. Use when the user asks to update Composer dependencies, audit packages for security vulnerabilities, manage composer.lock, configure Dependabot or Renovate for automated updates, replace abandoned packages, or resolve version conflicts.…
brainstorming
Interactive pre-implementation design exploration. Use when the user has a vague idea, feature request, or problem statement but no written spec. Asks focused questions one at a time to surface goals, non-goals, constraints, success criteria, and risks, then produces a written spec and hard-gates implementation until…