dotfiles

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.

Makes AI-generated content sound genuinely human — not just cleaned up, but alive. Use when content feels robotic, uses too many AI clichés, lacks personality, or reads like it was written by committee. Triggers: 'this sounds like AI', 'make it more human', 'add personality', 'it feels generic', 'sounds robotic', 'fix AI writing', 'inject our voice'.

Write, review, and improve posts for the Lightrains engineering blog in our specific voice. Use this for blog posts, technical articles, case studies, product announcements, deep dives, postmortems, and thought-leadership pieces aimed at engineers, founders, and CXOs.

Orchestrate a family of sub-skills to plan, design, and implement high-converting landing pages and homepage-like entry pages. Use when the user asks to create or redesign a landing page, campaign page, or homepage, even if they don't specify the page type or goal upfront.

Produce a low-fidelity wireframe and copy outline for each landing-page section, including layout, headlines, benefits, form fields, and microcopy. Use when the page structure is set and you need per-section content direction, or when the user asks for "copy," "content," or "wireframes."

Implement the approved landing page in code, reusing the project's stack, components, and design system. Use when the page plan is approved and needs to be built, or the user asks to "build," "code," "implement," or "generate" the page.

Clarify and structure a landing-page or homepage request into a precise, confirmed brief covering goal, audience, offer, traffic sources, and objections. Use when the user's request is vague, missing key details, or when the page's purpose and audience aren't clearly defined.

Read the current project to identify the tech stack, design system, reusable UI patterns, and constraints before planning a landing page. Use when connected to a codebase and the project context needs to be understood before deciding what to build or reuse.

Review the implemented page for UX, clarity, trust, and conversion, and suggest high-impact improvements. Use after a page is built to catch usability issues, optimize for conversion rate, or identify what to A/B test, even if the user only asks to "review" or "improve" the page.

Choose page type and plan section order using landing-page and homepage usability best practices. Use when the intent and project are understood but the section layout, page type (lead gen, sales, homepage), and above-the-fold content still need to be decided.

Run a lightweight accessibility audit on the landing page covering semantics, contrast, keyboard navigation, forms, and motion preferences. Use when the page code is ready and needs checked against WCAG-aligned heuristics, even if the user only asks to "check accessibility" or "make it accessible."

Map the approved wireframe to existing components, design tokens, and responsive behavior rules so the landing page feels native to the project. Use when the page structure is approved but the visual implementation — component selection, tokens, breakpoints — hasn't been decided.

Automate LinkedIn outreach safely with compliance-first practices. Handles connection requests, personalized messaging, and follow-up sequences while respecting LinkedIn's rate limits and detection systems. Triggers: 'message LinkedIn people', 'LinkedIn automation', 'send connection requests on LinkedIn', 'LinkedIn outreach', 'LinkedIn bot', 'message people on LinkedIn'.