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 working on Ruby code, gems, Rails-adjacent projects, Minitest or RSpec tests, Ruby packaging, Ruby style review, Ruby typing with RBI/RBS, modern Ruby syntax choices, or public API documentation with YARD.
Use when creating, placing, moving, reorganizing, or updating repository-local agent assets and artifacts that follow the dotagents layout under `.agents/`, including guides, skills, specs, tests, task state, logs, scratch files, or runtime state. Also use when deciding how `.agents/` relates to root project docs such as `PLAN.md`, `TODO.md`, `CHANGELOG.md`, `README.md`, and `AGENTS.md`.
Use when the user message starts with `:` to interpret prompt shortcuts such as `:commit`, `:co`, `:push`, `:pu`, `:ship`, `:sh`, `:opencode TEXT`, `:oc TEXT`, `:todo TEXT`, `:to TEXT`, `:ok NUMBER`, `:do NUMBER`, `:learn TEXT`, `:le TEXT`, `:close`, or `:cl`.
Use when working with Oracle Cloud Infrastructure via the OCI CLI, especially for session auth, budgets, quotas, networking, and compute operations. Read workspace AGENTS.md, if present, for workspace-specific tenancy and VPS state.
Use when editing, writing, or debugging Bash or POSIX shell scripts, shell-based automation, CLI wrappers, dotfiles, CI scripts, or when Bash style, safety, portability, or shell testing matter. Use when writing Bash code inside Markdown fenced code blocks. Always load this skill before writing or modifying any Bash code, even if the task seems straightforward — the skill catches subtle pitfalls that are easy to overlook.