Agent skill

Magic Hour Thumbnails

by hashgraph-online in hashgraph-online/awesome-codex-plugins

Create readable, truthful video thumbnails and covers with Magic Hour, preserving supplied faces and products and testing composition at small display size.

MITAuto-check passedMedia & Creative

Install Magic Hour Thumbnails

skills CLI
$ npx skills add hashgraph-online/awesome-codex-plugins --skill magic-hour-thumbnails -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install hashgraph-online/awesome-codex-plugins magic-hour-thumbnails --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/hashgraph-online/awesome-codex-plugins.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/magichourhq/skills/skills/magic-hour-thumbnails .claude/skills/magic-hour-thumbnails && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
magic-hour-thumbnails
GitHub stars
1.2k
Token cost
~1.2k tokens
SKILL.md length
603 words
Files
4 (incl. references)
Skills in repo
686
Repo updated
First seen
Licence
MIT

At a glance

Create readable, truthful video thumbnails and covers with Magic Hour, preserving supplied faces and products and testing composition at small display size.

  • YouTube thumbnails
  • SKILL.md covers Find the honest visual hook, Generate only the needed visual, Finish and judge small and Execute and retain
  • Reaches mcp.magichour.ai
  • Not a general product catalog set

What it does

Magic Hour Thumbnails is an agent skill from hashgraph-online/awesome-codex-plugins. Create readable, truthful video thumbnails and covers with Magic Hour, preserving supplied faces and products and testing composition at small display size. Use for YouTube thumbnails, Shorts covers, or episode covers, not a general product catalog set.

Its SKILL.md is about 1.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including reference files (for example `agents/openai.yaml` and `references/cover-design.md`).

It sits in Media & Creative. It works with YouTube. The repository describes itself as: A curated list of awesome OpenAI Codex / ChatGPT plugins, skills, and resources. The 1 Codex Marketplace. See live plugins at: https://hol.org/plugins/best-codex-plugins. The licence is MIT.

When your agent uses it

  • YouTube thumbnails
  • Not a general product catalog set

Example prompts

  • “/magic-hour-thumbnails”

What it can do on your machine

Read from SKILL.md and the folder at commit 78497e5. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • mcp.magichour.ai

    Also links to:

    • github.com
    • docs.magichour.ai

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Magic Hour Thumbnails loads about 1.2k tokens when it runs, and up to ~2k if it reads all its reference files. Until then it costs about 69 tokens; SKILL.md has 603 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~69
When it runs · the whole SKILL.md, loaded when a task matches
~1.2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from hashgraph-online/awesome-codex-plugins at commit 78497e5, republished under its MIT licence (© hashgraph-online). 603 words, ~1,181 tokens.

Download SKILL.mdSave it as .claude/skills/magic-hour-thumbnails/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
magic-hour-thumbnails
description
Create readable, truthful video thumbnails and covers with Magic Hour, preserving supplied faces and products and testing composition at small display size. Use for YouTube thumbnails, Shorts covers, or episode covers, not a general product catalog set.
license
MIT
metadata.author
magichourhq
metadata.version
1.1.0

Magic Hour thumbnails

Make the actual topic legible in a small image. Visual drama is useful only if the viewer can identify the subject and the promised content exists in the video.

Infer the hook from supplied content and recommend one composition; do not require a technical prompt or a long creative interview. Honor existing spending authorization and delegated creative decisions while respecting requested draft approvals. Show the finished cover first. For concept selection, editable headlines and series reuse, read cover design.

Find the honest visual hook

Read the supplied title, transcript, outline or footage. Choose one specific tension, comparison or outcome that it supports. Do not invent a result, reaction, endorsement, screenshot or person to make the cover more clickable. Use an authorized source image when face or product identity matters.

Build a composition around one dominant subject and one supporting cue. Put optional exact copy in a separate layout layer. Let copy complement the title rather than restating a long sentence; preserve supplied wording unless adaptation is authorized. Use the destination's actual aspect ratio and overlay regions; a thumbnail and a vertical cover need separately reviewed layouts.

Generate only the needed visual

Use ai_image_editor_create_image for reference-based changes; use ai_image_generator_create_image when no source must be preserved. State the subject's position, scale, expression if relevant, contrast and copy region. Request the change first and protected identity second. Do not depend on generated lettering for an exact headline.

Use my supplied workshop photo. Make the repaired blue chair the dominant subject on the right, with the damaged chair from my second photo smaller on the left. Preserve both chairs' actual construction and damage; do not exaggerate the result. Simple workshop background, clear separation, no added words. Leave the upper-left area clear for my exact headline.

If the brief supports two candidate concepts and the budget covers both, vary one meaningful idea: the comparison, framing, or subject expression. Do not spend on several nearly identical adjective changes. Generate the first candidate, inspect it, then decide whether the second answers a real question.

Show full SKILL.md (263 more words)Show less

Finish and judge small

Compose the exact headline/logo using available local layout tools and original assets. Keep an editable text source. Inspect the full-size export and a small preview near its actual feed size; around 160 pixels wide can expose hierarchy problems but is not a platform specification. Check recognition without zooming, text contrast, clutter, crop and overlays. Grayscale can reveal weak value separation, but does not replace color inspection.

If the subject disappears at small size, enlarge it or simplify the background before adding more text. If the exact copy clips, wrap or resize it; do not silently shorten it. If the face/product drifted, return to the original reference. Keep any watermark visible and disclose it. A visually preferred candidate is not a measured click-through winner.

Execute and retain

Use the creation MCP at https://mcp.magichour.ai/; connection guide. Read live models, input limits, resolution and prices. For API use, inspect the OpenAPI, then /v1/ai-image-editor or /v1/ai-image-generator. Stay inside existing spending authorization.

Upload local inputs through video_assets_generate_presigned_url, PUT raw bytes without the API bearer token to storage, and use file_path. Create once, retain the ID, then wait_for_image_project; a wait timeout means retrieve the same job. Do not repeat an ambiguous creation request. Preserve signed URLs and save the final file locally.

Deliver the cover, editable copy, dimensions and any rejected requirement. Reuse the accepted series palette, type hierarchy and crop rules for the next episode, while changing the visual hook to match that episode. The Nori case publishes two authenticated image edits and a finished layout; broader thumbnail quality and audience response remain unmeasured.

© hashgraph-online, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files (references) in plugins/magichourhq/skills/skills/magic-hour-thumbnails of hashgraph-online/awesome-codex-plugins.

  • SKILL.md
  • LICENSE
  • agents/openai.yaml
  • references/cover-design.md

Open the folder on GitHubat commit 78497e5

Compare with similar skills

Magic Hour Thumbnails next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Magic Hour Thumbnails compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Magic Hour Thumbnails this skillhashgraph-online/awesome-codex-plugins1.2k—~1.2kAutomated safety check: PassMIT
Native Subtitle Quote Imagechengyi-ai/native-subtitle-quote-image2.2k2 repos~1.8kAutomated safety check: PassMIT
Video Spec Builderfeicaiclub/video-spec-builder1k—~2.9kAutomated safety check: PassMIT
Wjs Uploading Videojianshuo/claude-skills1301 repos~2.1kAutomated safety check: PassMIT
Video Dataoxylabs/agent-skills875—~1.4kAutomated safety check: PassMIT
Motion Video in Remotionooiyeefei/ccc494—~2.9kAutomated safety check: PassMIT

Similar skills

  • Native Subtitle Quote Image

    chengyi-ai/native-subtitle-quote-image

    将本地视频或用户有权处理的在线视频,经过来源获取、文字稿定位、选题选句、精确取帧、紧凑裁切、拼图和逐张质检,制作成 3:4 或保留画面原比例的视频字幕长图。支持两种明确分开的输出:保留画面内已烧录字幕的原生字幕模式,以及把已审核的时间点与台词绘制到真实视频帧上的脚本字幕模式。用户要求原生字幕截图、字幕帧拼图、YouTube…

    2.2k GitHub starsUsed in 2 repos~1.8k tokens
    Media & CreativeAuto-check passed
  • Video Spec Builder

    feicaiclub/video-spec-builder

    当用户说想做一个视频、宣传片、产品演示、动画短片、抖音/YouTube 内容,或者说要改分镜、调节奏、换镜头、调字幕、加配音、改转场时使用。通过苏格拉底式追问收集视频需求,主动激发渲染层的全部能力(TTS / 字幕 / 3D / shader / 音频反应等),输出标准化的 video-spec.md 用于渲染。

    1k GitHub stars~2.9k tokensUpdated 4 mo ago
    Media & CreativeAuto-check passed
  • Wjs Uploading Video

    jianshuo/claude-skills

    Upload one or many videos to YouTube. An agent skill from jianshuo/claude-skills.

    130 GitHub starsUsed in 1 repo~2.1k tokens
    Media & CreativeAuto-check passed
  • Video Data

    oxylabs/agent-skills

    YouTube data extraction API and high-bandwidth proxy downloads.

    875 GitHub stars~1.4k tokensUpdated 8 days ago
    Media & CreativeAuto-check passed
  • Builds marketing and explainer videos in Remotion from rendered scenes, with one real product capture as proof, and cuts them for each platform's formats.

    494 GitHub stars~2.9k tokensUpdated 2 mo ago
    Media & CreativeAuto-check passed
  • Packaging Youtube Thumbnails

    ericosiu/ai-marketing-skills

    A skill your agent uses when a user supplies new video content or a channel and wants on-brand YouTube titles, thumbnail concepts, rendered variants, A/B packaging, identity profiling, precise…

    3.6k GitHub stars~3.8k tokensUpdated 16 days ago
    Media & CreativeAuto-check passed

More from hashgraph-online/awesome-codex-plugins

All 686 skills in this repo
  • Anime Reaction Gif

    hashgraph-online/awesome-codex-plugins

    Create original anime-style reaction stickers as looping GIFs and MP4 previews, using generated character pose sheets and timed key poses.

    1.2k GitHub stars~922 tokensUpdated yesterday
    Auto-check passed
  • Calibredb

    hashgraph-online/awesome-codex-plugins

    Manage and query Calibre libraries with the calibredb CLI (local paths or Calibre Content server URLs).

    1.2k GitHub stars~1k tokensUpdated yesterday
    Auto-check passed
  • Rust API Test Harness

    hashgraph-online/awesome-codex-plugins

    A skill your agent uses when adding, changing, testing, or debugging Rust HTTP APIs and services, especially when Codex needs black-box integration tests, random-port app startup, real database test…

    1.2k GitHub stars~1.7k tokensUpdated yesterday
    Auto-check passed
  • Art

    hashgraph-online/awesome-codex-plugins

    Make a studio's game look like something at build time — a cover from a real frame of the game (free), painted covers, backdrops, textures and character plates from image models through the…

    1.2k GitHub stars~2.4k tokensUpdated yesterday
    Auto-check passed
  • Game Balance Economy

    hashgraph-online/awesome-codex-plugins

    Balance game difficulty, resources, rewards, probability, progression, economies, and dominant strategies.

    1.2k GitHub stars~618 tokensUpdated yesterday
    Auto-check passed
  • Manuscript Engagement Analytics

    hashgraph-online/awesome-codex-plugins

    Analyze nonfiction manuscripts for reader engagement signals, including heading-level word counts, slow starts, long slogs, weak takeaway titles, value pacing, beta-reader comment dropoff, and…

    1.2k GitHub stars~875 tokensUpdated yesterday
    Auto-check passed

Works with

Questions about Magic Hour Thumbnails

What does Magic Hour Thumbnails do?

Create readable, truthful video thumbnails and covers with Magic Hour, preserving supplied faces and products and testing composition at small display size. Magic Hour Thumbnails is an agent skill from hashgraph-online/awesome-codex-plugins. Create readable, truthful video thumbnails and covers with Magic Hour, preserving supplied faces and products and testing composition at small display size.

When should I use Magic Hour Thumbnails?

Magic Hour Thumbnails fits situations like: youTube thumbnails; not a general product catalog set.

How do I install Magic Hour Thumbnails in Claude Code?

Run `npx skills add hashgraph-online/awesome-codex-plugins --skill magic-hour-thumbnails -a claude-code`. Or copy the skill folder (plugins/magichourhq/skills/skills/magic-hour-thumbnails in hashgraph-online/awesome-codex-plugins) into .claude/skills/magic-hour-thumbnails in your project. Claude Code loads it when a task matches its description.

How do I install Magic Hour Thumbnails in Codex?

Run `npx skills add hashgraph-online/awesome-codex-plugins --skill magic-hour-thumbnails -a codex`. Or copy the skill folder (plugins/magichourhq/skills/skills/magic-hour-thumbnails in hashgraph-online/awesome-codex-plugins) into .agents/skills/magic-hour-thumbnails in your project. Codex loads it when a task matches its description.

Can I use Magic Hour Thumbnails in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add hashgraph-online/awesome-codex-plugins --skill magic-hour-thumbnails -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/magic-hour-thumbnails, .gemini/skills/magic-hour-thumbnails, .github/skills/magic-hour-thumbnails and .opencode/skills/magic-hour-thumbnails in your project.

What does Magic Hour Thumbnails need to run?

SKILL.md names no scripts, command-line tools or credentials: Magic Hour Thumbnails is instructions for the agent only.

Does Magic Hour Thumbnails access the network?

SKILL.md names 3 domains. In commands or code: mcp.magichour.ai; the agent is likely to contact it when it follows the instructions. As links in the text: github.com and docs.magichour.ai. This is read from the text; nothing was executed.

Is Magic Hour Thumbnails safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Magic Hour Thumbnails use?

Magic Hour Thumbnails is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Magic Hour Thumbnails use?

About 1.2k tokens (SKILL.md is roughly 4.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 814 tokens, read only when the agent opens those files.

What are the alternatives to Magic Hour Thumbnails?

Skills that share tags, products or a category with Magic Hour Thumbnails: Native Subtitle Quote Image (chengyi-ai/native-subtitle-quote-image, 2.2k stars), Video Spec Builder (feicaiclub/video-spec-builder, 1k stars), Wjs Uploading Video (jianshuo/claude-skills, 130 stars) and Video Data (oxylabs/agent-skills, 875 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Magic Hour Thumbnails?

hashgraph-online (a GitHub organization) maintains it in hashgraph-online/awesome-codex-plugins, which has 1,242 GitHub stars. The repository holds 686 skills in this directory. The repository was last updated on October 8, 2026.

Source: hashgraph-online/awesome-codex-plugins on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.