By 46ki75
Org-internal engineering standards for this organization's projects.
My personal Claude Code plugin marketplace.
Each plugin lives under plugins/<name>/ with a .claude-plugin/plugin.json
manifest and bundles whatever it ships — Agent Skills, hooks, commands, or
agents. The marketplace itself is declared in .claude-plugin/marketplace.json.
Not intended for external contributors — feel free to fork or copy ideas, but issues/PRs aren't actively triaged.
/plugin marketplace add 46ki75/claude-plugins
/plugin install ai-protocols@46ki75-plugins
/plugin install wsl-notification@46ki75-plugins
| Plugin | Ships | Description |
|---|---|---|
ai-protocols | Skills | A2A, A2UI, AG-UI, and MCP protocol knowledge skills. |
qwik | Skills | Qwik (v1/v2) and Qwik City / Qwik Router. |
web-view-transition | Skills | The View Transition API (SPA and cross-document). |
wxt | Skills | Building browser extensions with WXT. |
prompt-evaluation-claude-code | Skills | Eval-driven prompt refinement run inside Claude Code. |
authoring | Skills | Markdown linting/fixing and Mermaid diagrams. |
conventional-commits | Skills | Conventional Commits messages. |
development-standards | Skills | Org-internal engineering standards. |
rust | Skills | Toasty ORM (v0.7) guidance. |
wsl-notification | Hooks | Windows toast notifications for Claude Code events from WSL2. |
plugins/ — published plugins, each a marketplace entry.skills/ — standalone Agent Skills that are intentionally not published
as marketplace plugins (e.g. kedb), plus a staging area for new skills
before they are bundled into a plugin.crates/ — Rust workspace that validates, archives, and publishes skill ZIPs
to the agentskills.io release channel..agents/skills/ — skills authored by other providers (reference only).submodules/ — upstream repositories tracked as git submodules.terraform/ — GitHub repo configuration (rulesets, labels).A plugin that bundles skills looks like:
plugins/ai-protocols/
├── .claude-plugin/plugin.json # manifest: name, version, description, …
├── README.md
└── skills/
└── <skill-name>/SKILL.md # auto-discovered by Claude Code
Claude Code auto-discovers skills/, commands/, agents/, and
hooks/hooks.json inside each plugin, so grouping skills into a plugin is just
a matter of placing their directories under skills/.
This repo was migrated from the former 46ki75/skills repository, where the
15 skills were distributed as agentskills.io ZIP releases. Thirteen are now
published as Claude Code plugins:
| Plugin | Bundled skills |
|---|---|
ai-protocols | a2a-knowledge, a2ui-knowledge, ag-ui-knowledge, mcp-knowledge |
qwik | qwik |
web-view-transition | web-view-transition |
wxt | wxt |
prompt-evaluation-claude-code | prompt-evaluation-claude-code |
authoring | markdown, mermaid |
conventional-commits | conventional-commits |
development-standards | development-standards |
rust | rust-toasty |
Two skills are kept standalone under skills/, intentionally not published as
plugins:
skills/kedb/ — maintains a local Known Error Database.skills/prompt-evaluation/ — eval-driven prompt refinement for the
Anthropic / Claude API (its Claude Code counterpart ships as the
prompt-evaluation-claude-code plugin).The skill-cli ZIP pipeline (under crates/) covers both channels: it
scans standalone skills under skills/* and plugin-bundled skills under
plugins/*/skills/*, so every skill is published as an
agent-skills-<name>-v<version> release regardless of where it lives. Because a
skill's name becomes its release tag, the scan errors out on any duplicate
name across the two channels.
# Validate every skill (skills/* and plugins/*/skills/*)
cargo run -p skill-cli -- check
# Markdown lint
just check
# Init submodules (run at repo root)
git submodule update --init --recursive
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub 46ki75/claude-plugins --plugin development-standardsSends Claude Code Notification and Stop events as Windows toast notifications from WSL2 via powershell.exe.
Expert knowledge skills for AI agent protocols: A2A (Agent2Agent), A2UI (Agent to UI), AG-UI (Agent–User Interaction), and MCP (Model Context Protocol).
Expert guidance for the Qwik framework (v1 and v2) and its meta-framework (Qwik City / Qwik Router).
Expert guidance for implementing the View Transition API — same-document (SPA) and cross-document transitions.
Rust development skills. Currently bundles expert guidance for the Toasty ORM (v0.7).
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Design fluency for frontend development. 1 skill with 23 commands (/impeccable polish, /impeccable audit, /impeccable critique, etc.) and curated anti-pattern detection.
Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations on LLM coding pitfalls