Explore skills

Resume Skill for Claude Code: An intelligent resume generator featuring ATS optimization, JD keyword matching, multi-format export, and 16 subject templates. It provides a local privacy-first resume workflow for Claude Code.

An anti-trope engine for AI writing, employing a core writing method (3 core types), 7 tools, 7 strict quality checks, and a "golden finger" technique system to ensure AI-written novels lack an "AI feel". Validated across 47 chapters, it's compatible with Claude Code, Cursor, and Windsurf.

Prepare a working local repo for a public GitHub release by auditing secrets and history, hardening setup and tests, writing a build-in-public README, packaging a clean zip, and publishing with public-ready metadata.

Turns vague work requests into 5-slot XML prompts by grilling the user on missing slots. Fire with /prompt-grill or "프롬프트 다듬어줘".

This skill allows users to check GitHub trending projects, compare daily, weekly, or monthly ranking changes, and generate trending reports.

X (Twitter) bookmark intelligence skill for Claude Code.

Generates a Claude-styled HTML 'story page' documenting the evolution of a vibe-coding project, from the user's initial idea to key design decisions, presented as a curated dialogue between the project owner and Claude.

Review a medical bill or EOB against insurance benefits to determine if charges are legitimate or disputable, and draft communications if needed.

awp-wallet session token; only legacy awp-wallet (<v0.17) needs this.

Analyzes music files (MP3, WAV, FLAC, M4A, OGG), providing a detailed music and lyrics analysis. This includes genre recognition, mood analysis, BPM, key, instrument detection, and lyrics interpretation, leveraging CUDA GPU acceleration for audio feature extraction.

Domain expertise for cell-cell communication (CCC) analysis, including intent-to-tool routing, workflow patterns, code templates, and pitfall warnings. Covers LIANA+, CellPhoneDB, CellChat, NicheNet, COMMOT, Squidpy, MEBOCOST, DIALOGUE, and 10+ other tools for scRNA-seq and spatial transcriptomics data.

Cost-efficient code editing uses a 4-tier cost ladder, automatically selecting the cheapest approach (LSP-scoped ast-grep, ast-grep, Haiku, or Sonnet) for any code modification across single files or entire codebases. It requires ast-grep and a language server.