SKILL.md packages that extend Claude Code, Cursor, Copilot, and other AI agents.
Tags
ai-helpers
Run a six-specialist reviewer panel on an Architectural Decision Record (ADR) and produce a consolidated PDF report and PPTX slide deck to surface risks, recomm
crucible
Adversarial review skill: iteratively attack artifacts (designs, plans, code, docs) and surface fatal/significant issues until clean or escalation.
humanizalo
Detects 40 AI writing ‘tells’ and rewrites text to inject a distinct human voice, with a 3-pass self-audit loop and 6‑dimension scoring.
slot-machine
Run N independent implementations of the same clear spec in parallel, review each, and pick or synthesize the best result to maximize quality when design choice
XHSSpec
Repo-first workflow for Xiaohongshu (XHS) content: plan campaigns, draft quick notes, evaluate trends, review drafts, and archive lessons as versioned Markdown/
capsem
Structured, per-bug review workflow: confirm, validate or push back, get agreement, implement TDD fixes, and commit with changelog — one bug per commit.
orchestrate
Supervisor skill that composes multi-model runs, routes subtasks to the right models and skills, and manages implementation & verification runs.
shuorenhua
Refine Chinese and English text to remove AI/template tone while preserving facts, terminology, and appropriate register for chat, docs, status, or public writi
proma
Execute a written implementation plan in safe, review-checked batches: load, review, run tasks in small batches with verification and checkpoints for feedback.
ryanb/dotfiles
Run the project's review tool (`bin/claude-review --print`), report the findings, and iteratively apply fixes, committing each change individually.
awesome-omni-skills
Audit screens and flows against Nielsen's heuristics and mobile UX best practices (StyleSeed Toss context), preserving upstream provenance and copied workflow f
swarms
Facilitator-focused procedure to close stuck review findings by requiring concrete evidence or concession, enforcing finality once decided.
Claude Code Skills
Merges and deduplicates evaluation worker results into a single verified review summary for coordinators, applying an architecture gate and verification checks.
Vern Bot
Persona-driven UX reviewer that evaluates features, journeys, and error states from an empathetic, user-first viewpoint to surface usability issues and fixes.
