agentleFS
jqaisystemsjqaisystems/jqai-ai-skills/skills/media-spec-checker/SKILL.md

media-spec-checker

Skill3 starsChanged 6 months agoMITView on GitHub

Check local image and video files against explicit technical requirements and produce measured pass, fail, or unverified results. Use for delivery-format checks, media-folder QA, dimensions, display ratios, duration, codecs, file sizes, or naming checks, not creative scoring or automatic conversion.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the rules contract. Adapt the fictional profile to the supplied specification.

    line 13
  2. Distinctive:

    Run the helper, using a finite timeout:

    line 14

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Media Spec Checker

Turn a supplied delivery specification into reproducible checks without changing the media.

Workflow

  1. Identify the exact local files and intended delivery requirements. Do not infer current platform limits from memory.
  2. Read the rules contract. Adapt the fictional profile to the supplied specification. Omitted checks are not performed.
  3. Run the helper, using a finite timeout:

    python scripts/check_media.py sample-media --rules rules.json --recursive --json
  4. Explain each failure and each unverified measurement. Missing tools, malformed media, unsupported files, absent durations, or an unconfigured media type are not passes.
  5. Return measured values, the requirements used, the report counts, and the smallest suggested correction. Do not convert or alter files unless a separate task authorizes it.

Dependencies and limits

Python 3.11 or newer. Image inspection requires Pillow; video inspection requires ffprobe on PATH. Explain missing dependencies without installing anything automatically. Animated image timing is measured when frame delays are available; static images have no duration. Display dimensions account for image orientation and right-angle video rotation; video display ratio also uses sample aspect ratio.

The timeout bounds each ffprobe process, not Pillow decoding. Image inspection uses Pillow's decompression-bomb protection; very large or many-frame inputs can still require substantial local work.

The checker inspects supported local containers, not playback quality, visual legibility, accessibility, licensing, or platform acceptance. It does not run network probes, re-encode media, or remove metadata. Unknown files in a selected folder remain visible as unverified.

Privacy and output

Reports contain selected measurements and relative filenames, not arbitrary embedded tags. Filenames can still identify projects; keep real reports private and use generated assets for public examples. Linked directories and files are not followed.

Exit codes: 0 means every requested check passed; 1 means measured failures; 2 means invalid input or any unverified result. The JSON counts preserve measured failures even when the overall exit is 2.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.