Agent skills
Skills you can use with AI coding agents, indexed from public GitHub repositories.
-
therapy-orchestrator
Master coordinator for NextStep Therapy's 12 skills. Diagnoses what the user needs, routes to appropriate skills, manages state across sessions, and handles context handoffs with Context Paradox awareness. Use when starting new projects, when user says 'help me with [X]' without specifics, when switching between tasks, or when determining which skill(s) to apply. Prevents context overload that degrades output quality. Includes 5 pre-built therapy workflows for common scenarios. Triggers on: help me with marketing, where do I start, what should I do first, I need to, complex multi-skill requests, ambiguous content requests.
majiayu000/claude-skill-registry 163
-
claude-artifact-creator
Creates, improves, and validates Claude Code artifacts (skills, agents, commands, hooks). Use when creating domain expertise skills, specialized task agents, user-invoked commands, extending Claude Code capabilities, improving existing artifacts, reviewing artifact quality, analyzing code for automation opportunities, or consolidating duplicate artifacts.
majiayu000/claude-skill-registry 163
-
Meta Prompt
Generate prompt variations from templates. Use when creating new commands, skills, or workflows.
majiayu000/claude-skill-registry 163
-
conductor
Context-driven development methodology for organized, spec-first coding. Use when:
- Project has a `conductor/` directory
- User mentions specs, plans, tracks, or context-driven development
- Files like `conductor/tracks.md`, `conductor/product.md`, `conductor/workflow.md` exist
- User asks about project status, implementation progress, or track management
- User wants to organize development work with TDD practices
- User invokes `/conductor-*` commands (setup, newtrack, implement, status, revert, validate, block, skip, revise, archive, export, refresh)
- User mentions documentation is outdated or wants to sync context with codebase changes
Interoperable with Gemini CLI extension and Claude Code commands.
Integrates with Beads for persistent task memory across sessions.
majiayu000/claude-skill-registry 163
-
browser-automation
Browser automation via Puppeteer MCP for JS-rendered content
majiayu000/claude-skill-registry 163
-
systematic-development
Use Sherpa's systematic development workflow for general coding tasks. Activates when starting development work, adding features, or implementing functionality. Ensures guide check before coding and guide done after completion for progress tracking and celebrations.
majiayu000/claude-skill-registry 163
-
Jira Ticket Estimator
This skill should be used when estimating development time for Jira tickets. It provides both manual and AI-assisted estimates with T-shirt sizes, story points, and phase-by-phase time breakdowns based on task type classification, complexity scoring, and project architecture (monolithic/serverless/frontend/fullstack/mobile/test_automation).
majiayu000/claude-skill-registry 163
-
clear-memory
Clean up old memory files
majiayu000/claude-skill-registry 163
-
n8n-automation
n8n workflow automation for building analytics including SkySpark multi-agent systems, FastAPI tool servers, workflow orchestration, and automated building system alert triage
majiayu000/claude-skill-registry 163
-
workflow-schema-composer
This skill should be used when the user wants to create a new looplia workflow, generate
a workflow definition file, or compose workflow steps from skill recommendations. Use when
someone says "create a looplia workflow", "generate workflow.md", "compose workflow steps",
"build me an automation pipeline", or "/build" (final step).
Final step in looplia workflow building: transforms skill recommendations into valid v0.6.3
workflow YAML/Markdown files. Each step uses skill: + mission: syntax, following the one
workflow step → one skill-executor → multiple skills architecture.
v0.6.3: Supports input-less workflows using input-less capable skills (e.g., search).
majiayu000/claude-skill-registry 163
-
claude-do-not-stop
Enable continuous work mode. Use PROACTIVELY when user says "keep working", "don't stop", "finish everything", "complete all tasks". Automatically continues until work is complete. User can invoke with /claude-do-not-stop [reason].
majiayu000/claude-skill-registry 163
-
debate-workflow
Structured multi-perspective debate for important architectural decisions and complex trade-offs
majiayu000/claude-skill-registry 163
-
session-docs
Load when closing beads tasks, ending significant sessions, or when planning discussions need to be captured
majiayu000/claude-skill-registry 163
-
beat-scheduling
Periodic task scheduling patterns with Celery Beat (crontab, interval, solar). Use when configuring periodic tasks, setting up task schedules, implementing recurring jobs, configuring django-celery-beat, or creating dynamic schedules.
majiayu000/claude-skill-registry 163
-
managing-github-ci
Configures GitHub Actions workflows and CI/CD pipelines. Manages automated releases via Changesets, PR validation, and Husky hooks. Troubleshoots CI failures. Triggers on: GitHub Actions, CI pipeline, workflow, release automation, Husky hooks, gh CLI, workflow failure.
majiayu000/claude-skill-registry 163
-
memory-graph
Persistent memory graph skill using the MCP Memory server
majiayu000/claude-skill-registry 163
-
confluence-to-agent-knowledge-base
Turn any Confluence Cloud instance into a private, searchable AI knowledge base on GitHub with access-tiered repos. Interactive 6-phase walkthrough using the Atlassian MCP connector (no API tokens needed). Ingests pages, compiles a structured wiki with categories and summaries, sets up search/Q&A CLI tools, splits into access-controlled GitHub repos, and configures weekly auto-sync. Use when asked to "migrate confluence", "create knowledge base from confluence", "set up a KB", "export confluence to github", "confluence to markdown", "build a knowledge base", or "make confluence searchable".
majiayu000/claude-skill-registry 163
-
n8n-v2
Comprehensive n8n v2.0 reference for building workflow automation. Use this skill when working with n8n workflows, especially for (1) creating workflow JSON files for import, (2) implementing human-in-the-loop approval workflows with Wait nodes, (3) troubleshooting Execute Sub-Workflow node issues, (4) setting up API credentials (LinkedIn, Anthropic, NewsAPI), (5) building orchestrator patterns with sub-workflows, (6) using expressions in nodes, (7) understanding MCP server limitations, or (8) any n8n workflow development task.
majiayu000/claude-skill-registry 163
-
email-marketing
Эксперт по email-маркетингу. Используй для email кампаний, автоматизации, сегментации, deliverability и A/B тестов.
majiayu000/claude-skill-registry 163
-
Project Memory Recall
Retrieve project-specific insights from file-based memory. Use when working on features, encountering domain-specific questions, or user says "--project-recall" or "--recall" (you decide which scope, may use both). Skip for routine tasks or universal pattern questions (use coder-memory-recall). MUST be invoked using Task tool to avoid polluting main context.
majiayu000/claude-skill-registry 163
-
task-skill
汎用的なタスクを実施する
majiayu000/claude-skill-registry 163
-
detecting-arbitrary-read
Detects arbitrary read vulnerabilities by identifying unchecked array indexing and out-of-bounds memory access. Use when analyzing array access patterns, pointer arithmetic, or investigating information disclosure vulnerabilities.
majiayu000/claude-skill-registry 163
-
xc-console
majiayu000/claude-skill-registry 163
-
humanizar-texto-es
Use when the user provides a text file (PDF, DOCX, MD, TXT) written in Spanish that was generated by AI and needs to be rewritten to sound naturally human-written, for research purposes on AI detection evasion techniques.
majiayu000/claude-skill-registry 163