Add historical cinema watches to a user's Letterboxd diary from receipts, booking confirmations, ticket text, screenshots, or pasted showtime lists. Use when Codex must parse film titles and screening dates, prepare one or more dated Letterboxd diary entries in the user's logged-in browser, avoid duplicates, obtain required save confirmation, and verify the resulting diary rows.
Create or refine optional blog post sketch images for carteakey.dev. Use this when a post needs a unique monochrome transparent illustration, when existing sketch assets need cleanup for feed/post rendering, or when wiring a post to the site-wide `image` / `imageAlt` workflow.
The author's specific writing fingerprint learned from reading their blog posts. AI writing tropes to avoid patterns that make output feel generated rather than written.
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.
Unified inbox ingestion skill for carteakey.dev. Scans inbox/ subdirectories, diffs against already-imported files, shows a preview of what will be imported and handles 4 distinct content targets: photography (photos.yaml), vibes (src/static/img/vibes/), ai-memes folio (folio/ai-memes/index.html), and loose root inbox files that need classification first.
Workflow for maintaining the inbox/manifest.yaml file. Use this to describe new files in the inbox and classify them before ingestion.
Ingests raw text, threads, or notes from the inbox/ directory, refines the tone, and formats them into standard markdown posts or notes.
Technical constraints, gotchas, and debugging tips specific to this site's Eleventy 3.0 configuration and data cascade.