Agent Skill
2/7/2026use-skill
Meta-skill that catalogs all available skills, explains when and how to use each one, and auto-discovers new skills. ALWAYS read this skill first before starting work to understand what tools are available. Use this as the single source of truth for skill selection.
Z
zenchantlive
0GitHub Stars
1Views
npx skills add zenchantlive/Stacean
SKILL.md
| Name | use-skill |
| Description | Meta-skill that catalogs all available skills, explains when and how to use each one, and auto-discovers new skills. ALWAYS read this skill first before starting work to understand what tools are available. Use this as the single source of truth for skill selection. |
name: use-skill description: "Meta-skill that catalogs all available skills, explains when and how to use each one, and auto-discovers new skills. ALWAYS read this skill first before starting work to understand what tools are available. Use this as the single source of truth for skill selection."
Use-Skill: Skill Catalog & Usage Guide
ALWAYS READ THIS SKILL FIRST before starting any work. This is your single source of truth for what skills are available and when to use them.
Quick Start
# Scan for all available skills
bash skills/use-skill/scripts/scan-skills.sh
Currently Installed Skills
š§ Core Agent Capabilities
| Skill | When to Use | Key Features |
|---|---|---|
| agent-browser | Headless browser automation outside OpenClaw's built-in browser | Navigate, click, fill forms, screenshots, PDFs, network interception, console inspection |
| codebase-cartographer | Understanding large codebases | Rust-based agentlens CLI, generates .agentlens/ docs, finds TODOs, symbols, dependencies |
| agent-swarm-workflow | Multi-agent parallel implementation | NTM + Agent Mail + BV coordination, exact prompts for swarm execution |
| council | Complex decisions needing multiple expert perspectives | Multi-persona deliberation (architect, analyst, security), 3-turn debate structure |
| autonomous-skill-orchestrator | Tasks requiring autonomous execution without babysitting | Freeze intent ā plan ā execute ā loop, strict guardrails |
š§ Development & Code Quality
| Skill | When to Use | Key Features |
|---|---|---|
| skill-creator | Creating new skills or updating existing ones | init_skill.py, package_skill.py, validation |
| memory-hygiene | Vector memory cleanup and optimization | Audit LanceDB, wipe/reseed, disable auto-capture |
| ai-cron-gen | Generating cron expressions from natural language | "Every Monday at 3pm" ā "0 15 * * 1" |
š Project Management
| Skill | When to Use | Key Features |
|---|---|---|
| beads-workflow | Task tracking with bd (beads) | Create, update, sync tasks with git |
| agent-mail | Multi-agent coordination via mail system | Inbox/outbox, file reservations, threaded messaging |
š External Integrations
| Skill | When to Use | Key Features |
|---|---|---|
| moltbook-interact | Reading from Moltbook agent social network | Browse posts, analyze (requires API credentials) |
Skill Selection Decision Tree
Starting a task?
āāā Need browser automation?
ā āāā Use: agent-browser
āāā Working with large codebase?
ā āāā Use: codebase-cartographer
āāā Complex decision with trade-offs?
ā āāā Use: council
āāā Want to run autonomously without questions?
ā āāā Use: autonomous-skill-orchestrator
āāā Multi-agent parallel work?
ā āāā Use: agent-swarm-workflow
āāā Creating/updating a skill?
ā āāā Use: skill-creator
āāā Memory issues or cleanup needed?
ā āāā Use: memory-hygiene
āāā Task tracking with beads?
āāā Use: beads-workflow
Auto-Discovery
This skill auto-discovers new skills added after its creation. Run the scan script to update the catalog:
bash skills/use-skill/scripts/scan-skills.sh
Important Notes
- Always check this skill first - New skills may have been added since last session
- Skills can be combined - Example: Use council to decide architecture, then autonomous-skill-orchestrator to implement
- Some skills need setup - Check each skill's Prerequisites section before first use
- Ping the user - Always announce when activating a skill: "Using skill: [name]"
Skills to Avoid (Redundant/Broken)
| Skill | Reason | Alternative |
|---|---|---|
| god-mode | CLI not installable, docs incomplete | Manual git/gh commands |
| agentlens (npm) | No CLI binary, guidance only | Use codebase-cartographer (Rust CLI) |
| chromadb-memory | Redundant - we use LanceDB | Use memory-hygiene |
| supermemory | Requires OpenAI API + external DB | Use memory-hygiene |
References
- Full skill details: See individual SKILL.md files in
skills/[name]/ - New skills: Check
skills/use-skill/scripts/scan-skills.shoutput
Skills Info
Original Name:use-skillAuthor:zenchantlive
Download