Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, passive voice, negative parallelisms, and filler phrases.
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes
Deploy and manage projects on Vercel using token-based authentication. Use when working with Vercel CLI using access tokens rather than interactive login, e.g. deploy to vercel, set up vercel, add environment variables to vercel.
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs. Use when the user asks to test, debug, screenshot, or automate a local web app.
Converts an existing Agent Skill to the steer framework, preserving its content and behavior: license triage, mapping hand-rolled machinery to steer components, rebuild with steer new, validation, and a measured comparison against the original. Use when the user asks to convert, port, migrate, or rebuild an existing skill on steer.
Builds an Agent Skill (SKILL.md) end to end with steer: design the trigger and components, scaffold with steer new, write the body, then gate on steer validate and a real run. Use when the user wants to create, write, or scaffold a skill, turn a document, API, or repeated workflow into a skill, or review, fix, or improve an existing skill. Also use when a skill needs credentials, persistent state, enforced steps, background processes, or learning wired in.
Generates a repository health report: commit activity, code inventory, TODO debt, test presence. Use when the user asks for a repo health check, codebase overview, or tech-debt snapshot.
Writes a conventional commit message from the staged diff, one logical change at a time.