Media skills

Create interactive holographic cards from supplied images with colored layers, repaired scenery, signed parallax and contour glow. Use Codex native image tools with local assembly, or an explicitly selected independent API workflow.

The English summary is being prepared. Showing the original description.

Convert image references into spatial hypotheses, camera constraints, construction routes, material evidence, and measurable Blender reconstruction gates. Use whenever a user asks to replicate, recreate, match, clone, or reconstruct a scene, environment, object, material, lighting setup, or shot from one or more images, especially when room connectivity, hidden space, perspective, visible-frame fidelity, material identity, aging, wetness, or photographic treatment matters.

The English summary is being prepared. Showing the original description.

Analyze reference-image illumination and build accountable Blender lighting from shadow direction, shadow length, softness, exposed faces, environment fill, bounce, practical sources, and material response. Use for reference matching, daylight, interiors, exterior environments, product scenes, atmospheric lighting, or any task where lights must be physically and visually justified instead of added as arbitrary fill or rim lights.

The English summary is being prepared. Showing the original description.

Analyze, summarize, and extract insights from DeLive transcription sessions. Use when: user mentions DeLive, transcription, meeting transcripts, live captions, audio transcription, AI correction, corrected transcript, or transcript analysis; user wants to search, retrieve, summarize, correct, or process recorded transcripts; user asks about meeting notes, action items, discussion summaries, or transcript quality from DeLive. Requires DeLive app running locally with its MCP server or REST API.

The English summary is being prepared. Showing the original description.

Use when implementing video playback with controls.

The English summary is being prepared. Showing the original description.

Use when you need to display and browse image collections.

The English summary is being prepared. Showing the original description.

Use when you need to upload and preview images.

The English summary is being prepared. Showing the original description.

Fix SQLite "database disk image is malformed" errors caused by corrupted FTS5 indexes. Use when: (1) UPDATE/INSERT fails with "stepping, database disk image is malformed" but PRAGMA integrity_check returns "ok", (2) Only specific records fail to update while others succeed, (3) Records contain dirty text with tabs, newlines, or unusual characters that were inserted into FTS5-indexed columns. The misleading error message suggests disk corruption, but the actual issue is FTS5 index corruption from malformed text content.

The English summary is being prepared. Showing the original description.

Recover Vine video metadata (author, user ID, title, thumbnail) from vine.co oembed endpoint even though the Vine API and pages are dead. Use when: (1) you have Vine shortcode URLs (vine.co/v/XXXXX) and need to identify the creator, (2) the Vine page only shows a loading spinner, (3) archive-api.vineapp.com returns 404, (4) need to map Vine shortcodes to user IDs for cross-referencing against the divine.video database. The oembed endpoint is a still-live metadata oracle for the otherwise-dead Vine platform.

The English summary is being prepared. Showing the original description.

Fix "Failed to load video" errors when network requests succeed (200/206 status). Use when: (1) Video element shows error but DevTools shows successful requests, (2) Videos download correctly via curl but fail in browser, (3) Range requests work but video won't play, (4) Building CDN/media server for video streaming. The root cause is often missing Accept-Ranges header which browsers need for video seeking and streaming.

The English summary is being prepared. Showing the original description.

Fix iOS AVPlayer "CoreMediaErrorDomain error -12939 - byte range length mismatch" or similar video streaming failures when a proxy/edge/CDN service sits between the client and object storage (GCS, S3, R2). Use when: (1) iOS video fails with -12939 "byte range length mismatch - should be length 2 is length N", (2) curl with Range: bytes=0-1 returns full file instead of 2 bytes, (3) one URL path works with Range but another path to the same storage doesn't, (4) proxy advertises Accept-Ranges: bytes but returns 200 with full file for range requests. The root cause is typically a proxy handler that constructs a new request to the backend without forwarding the client's Range header.

The English summary is being prepared. Showing the original description.

Fix NIP-46 nostrconnect:// URLs not being recognized by signer apps (Amber, nsec.app, nsecBunker). Use when: (1) QR code or URL paste fails in signer app despite valid-looking URL, (2) Using metadata JSON object instead of separate query params, (3) Implementing client-initiated NIP-46 connections. The NIP-46 spec requires name/url/image as separate query parameters, not a metadata JSON object.

The English summary is being prepared. Showing the original description.

Fix duplicate Nostr events (Kind 30000+ parameterized replaceable events) caused by missing d-tag. Use when: (1) Same content appears multiple times on relay for same pubkey, (2) Events aren't being replaced/updated as expected, (3) Publishing NIP-71 video events (Kind 34236) or other addressable events and seeing duplicates. The d-tag is REQUIRED for addressability - without it, each publish creates a new non-replaceable event.

The English summary is being prepared. Showing the original description.

Fix "count shows X but list shows empty" bugs when querying related events (comments, reactions, zaps, reposts) for Nostr addressable events (Kind 30000-39999). Use when: (1) REST API shows engagement count > 0 but WebSocket query returns empty, (2) Comments/reactions exist but app shows "No comments yet" or similar, (3) Working with NIP-22 comments, NIP-25 reactions, or NIP-18 reposts on Kind 34235/34236 video events or other addressable events. Root cause: clients may reference addressable events using either E tag (event ID) or A tag (kind:pubkey:d-tag format), and querying only one tag misses events tagged with the other.

The English summary is being prepared. Showing the original description.

Fix "blocked: event rejected by relay policy" errors when publishing Kind 34236 (video) events to divine relays. Use when: (1) Kind 0 profile events succeed but Kind 34236 fails, (2) Generic policy rejection with no specific reason, (3) Video events work on other relays but fail on divine relays. The divine relay requires thumbnails for all video events - stricter than NIP-71 spec.

The English summary is being prepared. Showing the original description.

Fix "setState() or markNeedsBuild() called during build" error triggered from deactivate(). Use when: (1) Error occurs during widget disposal or navigation away, (2) Stack trace shows deactivate -> some notifier -> ValueListenableBuilder._valueChanged, (3) Video/audio player pause() or stop() calls in deactivate() cause the error. The fix is deferring state-modifying operations to addPostFrameCallback.

The English summary is being prepared. Showing the original description.

Fix silent video/media processing failures caused by URL extraction code that filters on file extensions (.mp4, .webm, .webp). Use when: (1) Media moderation, transcoding, or analysis silently skips files from Blossom or content-addressed storage servers, (2) URL extraction from Nostr event tags (imeta, r tags) drops URLs without recognized extensions, (3) CDN fallback URLs append .mp4 but the actual server uses extensionless content-addressed paths like /{sha256}. Common in Nostr video events (kind 34236) where different clients use different URL formats.

The English summary is being prepared. Showing the original description.

Fix duplicate Nostr events in ClickHouse when using ReplacingMergeTree for NIP-33 addressable events (Kind 30000+). Use when: (1) Edited videos/events appear as duplicates, (2) Same d_tag shows multiple events with different IDs, (3) FINAL keyword doesn't deduplicate properly for parameterized replaceable events. The issue is that FINAL deduplicates by ORDER BY key (typically `id`), not by (pubkey, kind, d_tag).

The English summary is being prepared. Showing the original description.

Generate diagrams, flowcharts, and conceptual illustrations using OpenAI gpt-image-2 (default) or gpt-image-1. Use for: algorithm/code flow diagrams, system architecture sketches, conceptual illustrations, and any figure that does NOT represent actual code execution output (data plots, model results, etc. must still be generated by running Python code). Requires OPENAI_API_KEY and OpenAI organization verification when used from Claude Code / API path. OpenAI Codex subscribers can use built-in $imagegen without an API key.

The English summary is being prepared. Showing the original description.

Use A4 Eight Page Booklet 2.2 to create, write, repair, impose, and validate a printable A4 one-sheet eight-page mini book or zine from text, notes, images, BIRD Graph JSON, iMandalArt/九宮, FIRE, outlines, templates, or PDFs. Use whenever the user says A4八頁小書, 八頁小書, 八頁小冊, 八格小書, A4一張八頁, 八頁在一頁, 8頁在一頁, 八張一頁, 一張紙變八頁, 摺頁書, 折頁小書, 迷你書, 小誌, zine, mini zine, booklet, 做成小書, 印成小書, 單張印出八頁, 幫我拼版, 八頁PDF, 合併八頁, 6-7-8-1／5-4-3-2排列, or asks for eight upright page PDFs, an eight-page reading PDF, an editable DOCX, or a correctly rotated A4 landscape print sheet.

The English summary is being prepared. Showing the original description.

Extract subtitles or a transcript from a YouTube URL and save normalized timestamped text locally. Use when the user wants the spoken content of a YouTube video as text.

The English summary is being prepared. Showing the original description.

Generate, edit, or composite images with Gemini/Nanobanana. Default for image requests without a named provider; do not override a different provider the user chose.

The English summary is being prepared. Showing the original description.

Generate images when the user names MiniMax or image-01/image-01-live. Use the default image skill when no provider is named.

The English summary is being prepared. Showing the original description.

Generate or edit images when the user names OpenAI, GPT Image, or a gpt-image model. Use the default image skill when no provider is named.

The English summary is being prepared. Showing the original description.