Turn a raw commercial offer (client + line items + total) into an HTML proposal whose style copies a brand website. LLM-authored from a brand screenshot; --quick offline fallback. Output: proposal.html with clickable links + exact prices, prints to PDF. Use when: 'make a proposal', 'commercial offer', 'КП'…
Turn a topic / research brief / script into a 9:16 vertical reel: 1-4 shots + matched music + ffmpeg-stitched final.mp4 with optional burned-in captions. Modes: --topic / --research / --script-file; --shots; --style; --music-style; --aspect; --captions. Use when: 'make a reel about X', 'short video on Y', 'TikTok /…
Write user-facing release notes + changelogs. Keep-a-Changelog format, sections Added/Changed/Fixed/Deprecated/Removed/Security. Per-audience tone (user/dev/ops). Anti-marketing-fluff bans. Wraps writer. Use when the user says 'release notes for vX.Y.Z', 'changelog entry', 'what's new', 'product update'.
Produce a structured research brief on any topic — TL;DR, cited key facts, notable quotes, suggested angles, open questions. 3-15 queries by depth, multi-source (WebSearch + WebFetch + Firecrawl/Exa MCP). Markdown for downstream skills. Use when: 'research X', 'gather facts on Y', 'brief me on Z', 'prepare research…
Generate JSON-LD structured data (Article, FAQPage, HowTo, Organization, Person with knowsAbout) and llms.txt from a markdown post. Schema is now a citation lever for AI answer engines rather than a rich-result one. Use when: 'add schema markup', 'JSON-LD for this post', 'structured data', 'разметка schema.org'…
Manage the local style library — bundled (24 carousel + 12 video director + 12 music genre) plus user overrides at /.claude/style-library/. CRUD + submit (upstream-PR packages) + schema validation. Use when: 'add a custom carousel style', 'create a directorial style', 'genre preset for k-pop', 'list available styles'…
Check if a newer Mikefluff/skills release exists; show CHANGELOG diff; on user confirmation, run install.sh --update. Idempotent and read-only until user approves. Use when the user says 'check for skill updates', 'update skills', '/skills-update', or sees the status-line update banner.
Read-only pre-commit lint over writer/prose-edit/essay-write. Takes a staged diff (or a specified file / commit range) and flags neuro-slop, synthetic structures, voice drift in prose. Read-only — never edits. Use before git commit, as a post-commit check, or as a pre-commit hook.
Visual style generator — turn a text description and/or reference image into a structured style entry for the prompt-library. Duplicate-detect then emit v2.15.0 schema (background, accents, mood, typography, compositionsignature). Use when: 'make a new style', 'add a style based on this image', 'добавь стиль'…
Apply an artistic style to an existing image. Default Flux Kontext. 12 presets: watercolor, oil, sketch, line-art, ink-wash, cyberpunk, ghibli, pixar-3d, manga, art-deco, low-poly, vaporwave. Custom via --prompt-mod. Use when: 'style transfer', 'make this look like X', 'стилизуй фото', 'переведи в стиль X', 'переведи…
Burn captions / subtitles onto a video via ffmpeg. Supports SRT, WebVTT, plain text. Presets: modern (white on black lower-third), minimal, bold (yellow, denser). Subcommands: burn, preview. Pure ffmpeg. Use when: 'add subtitles to my video', 'burn captions onto this MP4', 'subtitle my reel', 'добавь субтитры к…
Produce YouTube / blog / podcast-episode thumbnails. 16:9 default (1280×720, 1920×1080). Face + bold title aesthetic; face-placement variants (left / right / center). Model picked for face preserve + text rendering. Use when: 'YouTube thumbnail', 'blog header image', 'podcast episode cover', 'обложка для youtube видео…
Rewrite text in a different register without changing meaning — formal↔casual, business↔academic, technical↔friendly, plain-explainer. 6 named registers + transformation deltas. Wraps writer. Use when the user says 'make this more casual / formal / accessible / business-like', 'rewrite for younger audience'.
Transcribe audio / video to SRT / WebVTT / JSON / plain text via OpenAI Whisper or the GPT-4o transcribe models. Auto-detects language or accepts --lang ISO-639-1. $0.003-0.006/min. Pairs with subtitle-burner. API limit 25 MB/call. Use when: 'transcribe this video', 'subtitles from audio', 'speech to text', 'распознай…
Read-only parity checker for multilingual translations (RU↔EN↔PT-BR). Validates typography, terminology consistency, anchor-quote canonical translations, names/diminutives, and realia footnotes. Produces a parity report; never edits. Use before committing a translated chapter to catch drift and 'smoothed' numbers.
A prose editor and style checker for Russian or English text. It cleans writing, or reports style problems with their locations, using rules for typography, repetition, sentence structure, and unnatural wording.