Install and configure Azure CLI on macOS for local development. Use when the user needs az CLI installed, needs to log in to Azure, needs to check Azure auth status, or needs to manage subscriptions and access tokens.
Structure code so that the set of representable states equals the set of valid states for a given domain, resolving the boundary condition d(S) → ∅. Each domain is a type boundary; closing it adds a constraint layer. Domains have prerequisites — valid(Dn) requires closed(Dm) for all prerequisites Dm. Use when designing types, schemas, state machines, or any structure where invalid states should be unrepresentable.
Detect when code changes reopen closed domains or widen boundary conditions — expanding d(S) where it was resolved. Uses Hamming distance to measure change locality and delta-epsilon to characterize the exact boundary where a change crosses into another semantic domain. Respects the domain partial order: drift in a prerequisite domain invalidates dependent domains. Use when reviewing diffs, auditing generated code, or verifying refactors.
Constrain AI code generation to preserve or resolve boundary conditions — never expand them. Controls what is representable in the generation context rather than relying on instructions. The prompt defines the open interior, seeds shape continuation, types enforce closure, and delta-epsilon characterizes the generation boundary. Respects the domain partial order: generation context must reflect which domains are closed and which boundary conditions remain open.
Characterize the stochastic noise envelope of LLM code generation to distinguish acceptable sampling variance from semantic drift. Use when evaluating whether differences between generated outputs are noise or signal, establishing reproducibility criteria for generation tasks, determining confidence levels for context closure states, or calibrating ε thresholds for drift detection. Provides the probabilistic foundation that makes delta-epsilon analysis rigorous over a non-deterministic generator.
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.
On-demand skills for AI agents. Instead of stuffing every capability into the context window, agents discover and load only what they need — via a 2-tool MCP server that scales to any number of skills at constant cost.
Works with Claude, Copilot, and any MCP-compatible client.
Add the MCP server to any client:
{
"mcpServers": {
"skill-router": {
"command": "npx",
"args": ["-y", "@stuffbucket/skills"]
}
}
}
Or install as a plugin marketplace:
# Claude Code
/plugin marketplace add stuffbucket/skills
# Copilot CLI
/plugin marketplace add stuffbucket/skills
That's it. Your agent can now discover and load skills on demand.
The skill-router exposes two tools — list_skills and get_skill — instead of registering every skill separately. Agents search by intent, load what they need, and the context window cost stays
constant.
Each skill is a self-contained package: a SKILL.md with instructions plus optional scripts/, references/, and assets/. Skills are authored once by an expert and delivered automatically to
anyone whose agent needs them.
Anyone can contribute a skill. The entire workflow is scaffold → edit → validate → submit:
# 1. Scaffold
npm run new -- my-new-skill
# 2. Edit the generated SKILL.md
# - Set `name` and `description` in the frontmatter (this is what triggers the skill)
# - Write instructions in the body (what the agent needs to know)
# - Add scripts/, references/, or assets/ if needed
# 3. Validate
npm run validate:one -- plugins/stuffbucket/skills/my-new-skill
# 4. Package and submit
npm run package -- plugins/stuffbucket/skills/my-new-skill
# Open a "New skill" issue and attach the .skill file — CI validates and opens a PR
Key principles (from the skill-creator guide):
references/ files.See CONTRIBUTING.md for the full guide, or read the Best Practices.
| Skill | Description |
|---|---|
skill-router | MCP server — find and load skills on demand (context window optimization) |
skill-creator | Guide for creating, validating, and packaging new skills |
figma-make-to-vite | Integrate Figma Make exports into Vite + React + TypeScript projects |
file-management-skill | File and directory operations |
code-analysis-skill | Code quality and pattern analysis |
git-workflow-skill | Git operations (commit, branch, merge, rebase) |
testing-skill | Unit and integration test writing and execution |
pages-prepare-vite | Prepare a Vite project for GitHub Pages deployment |
pages-build-vite | Build and verify a Vite project for Pages |
pages-commit-vite | Stage and commit Pages configuration files |
pages-push-vite | Push branch to trigger GitHub Actions deployment |
pages-publish-vite | Monitor Actions deployment and report the live URL |
example-skill | Demonstrates proper structure and format |
npm run setup # install deps, create MCP symlinks, build skill index
| Command | What it does |
|---|---|
npm run setup | Install deps + create MCP symlinks + build index |
npm run build:index | Rebuild the skill-router index |
npm test | Build index + run MCP server smoke tests |
npm run test:repl | Build index + launch interactive REPL |
npm run lint | Lint markdown + JS + JSON schemas |
npm run validate | Validate all skills (frontmatter + content completeness) |
npm run validate:one -- <path> | Validate a single skill |
npm run new -- <name> | Scaffold a new skill |
npm run build:llms | Regenerate llms.txt from template |
npx claudepluginhub stuffbucket/skills --plugin stuffbucketThe complete library of Agent Skills — AI agents & orchestration, MCP & SDKs, Oracle & cloud, frontend & animation, engineering workflow, content & brand, and creative production. Skills auto-activate when relevant.
Skill discovery and installation assistant — find and install the right skills for your project
Create and validate production-grade agent skills with 100-point marketplace grading
Create and manage Claude Code skills, plugins, subagents, and hooks. Use when building new skills, validating existing skills, testing skills empirically, creating plugins, converting projects to plugins, creating hooks, or managing plugin automation. Includes /skills-toolkit:skill-composer, /skills-toolkit:skill-refiner, /skills-toolkit:skill-tester, /skills-toolkit:plugin-creator, /skills-toolkit:subagent-creator, /skills-toolkit:hook-creator, and /skills-toolkit:ask-user-question skills.
Skill synchronization and management for Claude Code, Codex, GitHub Copilot, and Cursor. Provides 27 MCP tools for validation, sync, intelligence, and tracing.
Plugin-safe Claude Code distribution of Antigravity Awesome Skills with 1,561 supported skills.