Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add mertbuilds/skills --skill feed-safe-imagesgit clone --depth 1 https://github.com/mertbuilds/skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mertbuilds/skills/feed-safe-images)<a href="https://agentmods.dev/skills/mertbuilds/skills/feed-safe-images"><img src="https://agentmods.dev/badge/skills/mertbuilds/skills/feed-safe-images/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mertbuilds/skills/feed-safe-images"><img src="https://agentmods.dev/badge/skills/mertbuilds/skills/feed-safe-images.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00105 | $0.02224 |
| Opus 5 | $0.00053 | $0.01112 |
| Sonnet 5 | $0.00021 | $0.00445 |
| Haiku 4.5 | $0.00011 | $0.00222 |
Grade A, and why
feed-safe-images scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Instagram's Graph API has no upload endpoint: `create_media_container` takes an `image_url`, and Meta fetches it once. `scripts/host.sh <file>...` uploads each file to a Cloudflare R2 bucket and prints one public URL per How it starts
The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Feed-safe images
Social feeds crop photos to a fixed ratio in the timeline. Only the full-screen view shows the whole frame, and most people never tap. The fix is to put the photo on a solid canvas at the ratio the feed will show, so the crop hits the padding instead of the photo.
This skill ships scripts/pad.sh, an ffmpeg one-liner wrapped with the right canvas per platform layout.
Ratios
| Layout | Platform and case | Canvas per image | Ratio |
|---|---|---|---|
x1 |
X, single image, photo at 95% width | 2400 x 1920 | 5:4 |
xc |
X, every slide of a 2-4 image carousel, photo at 95% width | 2400 x 1920 | 5:4 |
x2 x3big x3small x4 |
X, legacy grid layouts. Obsolete, see note below | ||
ig |
Instagram feed post or carousel slide, photo at 95% width | 2160 x 2880 | 3:4 |
igblack |
Instagram, same canvas as ig but black, photo corners rounded (91 px). The uncropped companion to film for horizontal photos |
2160 x 2880 | 3:4 |
film |
Instagram, single image or carousel slide, film-scan look: photo full-bleed (cover-cropped to 3:4) with a thin black border and rounded inner corners. Instagram only, X crops it (5:4 slot) | 1512 x 2016 | 3:4 |
threads |
Threads post or carousel slide | 2160 x 2700 | 4:5 |
li |
LinkedIn, portrait (most feed space) | 2160 x 2700 | 4:5 |
li-wide |
LinkedIn, landscape | 2400 x 1256 | 1.91:1 |
square |
any platform, 1:1 | 2160 x 2160 | 1:1 |
Canvases are 2x to 3x the platform's display size so they survive recompression. The photo is fit to 95.3% of the canvas on its constraining axis and centered (95% on x1, xc and ig, which leaves 2.5% side margins: in the X and Instagram feeds 95.3% read as zero margin, and 90% or 92% read as too-thick margins, both live-tested). The reference for the thinner look was a photographer's Instagram post at 1512 x 2016 with about 2.4% side margins; override with --fit.
X multi-image posts are a carousel now, not a grid. Since 2026 (X Lite Android April, wider rollout by June, x.com web by late August) a post with 2-4 images shows as a horizontal swipe carousel. Each slide renders at its own full aspect ratio with no crop, so the old grid pads (7:8, 4:7, 2:1) are wrong: they show as squat white-barred slides. The carousel slot itself is 5:4 with object-fit: cover (measured on x.com 2026-08-29: 479 x 383 at every viewport width), so anything wider than 5:4 is cropped (16:9 loses about 30% of its width) while portrait down to 7:8 passes through. Pad every slide with xc (5:4). x1 is 5:4 too now (it was 16:9): a 3:2 or 4:3 photo on a 16:9 canvas came out height-limited at about 76% width with fat side margins, and 5:4 makes it width-limited with bands top and bottom, the same look as the carousel. The x2 / x3big / x3small / x4 layouts stay in the script for anyone still seeing the grid, but do not reach for them by default.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago Changed · +10 lines f2a12929fb10
- 4d ago Changed · +10 lines · +13 tokens per session e64c5e928d26
- 12d ago First seen · 63 lines · 92 tokens per session scan A 9662591733bc
feed-safe-images is a skill published in the GitHub repository mertbuilds/skills (2 stars, last pushed 3d ago), licensed MIT. It adds 105 tokens to every session and 2,224 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…