huhhb

huhhb

huhhb is Uhstray.io's Claude Code skills marketplace — a curated library of skills (slash commands, workflows, AI automation) for teams building with Claude Code.

Name is a play on "hub" (a place for skills) + "uhh" (Uhstray). Say it fast.

---

Install

# 1. Add the marketplace
claude plugin marketplace add uhstray-io/huhhb

# 2. Install (user scope — available in all projects)
claude plugin install --scope user huhhb

# Or project-scoped — only in this repo
claude plugin install --scope project huhhb

memory requires MemPalace for its MCP server. Install it once, then the plugin uses it automatically:

uv tool install mempalace   # recommended
# or: pip install mempalace

---

Update

# Fetch latest from all marketplaces
claude plugin marketplace update

# Uninstall then reinstall — "install" silently skips if already present
claude plugin uninstall huhhb
claude plugin install --scope user huhhb

---

Add a New Skill

1. Create skills/<skill-name>/SKILL.md with YAML frontmatter:

---
name: skill-name
description: Use when [specific triggering conditions] — embed trigger phrases here, not in a triggers field
---

# Skill content here

Note: The triggers frontmatter field is not supported by VS Code agents. Put trigger phrases in the description instead.

2. Register it in marketplace.json:

{
  "name": "skill-name",
  "path": "skills/skill-name/SKILL.md",
  "description": "...",
  "category": "category",
  "tags": ["tag1", "tag2"],
  "version": "0.1.0"
}

3. Open a PR — see CONTEXT.md for quality requirements.

---

What's Inside

Onboarding

Skill	Trigger	Purpose
onboarding	/onboarding	First-time setup wizard — configure Auto Mode, enable Agent Teams, and orient to Claude Code

Memory

Skill	Trigger	Purpose
memory	/memory	Team memory nexus (Uhstray.io) — store and recall project context via MCP
memory-search	/memory-search	Semantic search across the nexus
memory-mine	/memory-mine	Ingest a project directory into the nexus
memory-status	/memory-status	Nexus stats — drawer count, wings, rooms
repo-memory	/repo-memory	Repo-local memory in .claude/memory/ — committed to git, no external service

Dev Workflows

Skill	Trigger	Purpose
using-superpowers	/using-superpowers	How to find and invoke skills — read this at session start if skills aren't auto-matching
brainstorming	/brainstorming	Explore intent and design before touching any code — required before implementing features
writing-plans	/writing-plans	Write a structured implementation plan from a spec or requirements
executing-plans	/executing-plans	Execute a written plan in a focused session with review checkpoints
test-driven-development	/test-driven-development	Write failing tests before writing implementation code
systematic-debugging	/systematic-debugging	Root-cause a bug before proposing any fix
verification-before-completion	/verification-before-completion	Run verification commands and confirm output before claiming work is done
grounding	/grounding	Pause a long session for a checkpoint — surfaces in-flight work, runs /simplify + /security-review and test/build/lint health, checks the diff against repo conventions, and re-confirms goals. Hook-fired (default 2h) or manual; opt-in
subagent-driven-development	/subagent-driven-development	Execute a multi-step plan using parallel subagents in the current session
dispatching-parallel-agents	/dispatching-parallel-agents	Split 2+ fully independent tasks across separate agents simultaneously
using-git-worktrees	/using-git-worktrees	Isolate feature work in a separate git worktree to avoid conflicting with the current workspace
finishing-a-development-branch	/finishing-a-development-branch	Decide how to integrate completed work — merge, PR, squash, or clean up
writing-skills	/writing-skills	Author, edit, or validate a new huhhb skill before shipping it

Explanation

Skill	Trigger	Purpose
explaining-plans	/explaining-plans	Augment a plan/spec/RFC in place — decision criteria, cited sources, target outcome, mermaid diagrams. Composes with writing-plans.
explaining-changes	/explaining-changes	Narrate changes as they happen — each change, each task, and before every commit. Brief prose + ASCII diagrams, chat-only.

Review