generate-consistent-og-images

generate-consistent-og-images is a skill for Claude Code, Codex from lossless-group/lossless-agent-skills. It costs 432 tokens per session (8,385 once invoked), scanned B, original, MIT.

A guide for creating OpenGraph images, the preview images shown when a webpage link is shared online. It supports coordinated banners, portraits, square images, and tall cards by either generating illustrations or rendering the site's own visual component.

In plain words
What is it for?
Use it to create or refresh link-preview images for websites, splash pages, plugins, or fundraising presentations, including formats for services such as WhatsApp and iMessage.
Why use it?
Shared links can look missing, outdated, or unrelated to the page. This guide helps keep preview images consistent with the site's design and available in the needed shapes.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to create or refresh link-preview images for websites, splash pages, plugins, or fundraising presentations, including formats for services such as WhatsApp and iMessage.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/lossless-group/lossless-agent-skills/generate-consistent-og-images
Install

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.

Any agent
npx skills add lossless-group/lossless-agent-skills --skill generate-consistent-og-images
Clone the repo
git clone --depth 1 https://github.com/lossless-group/lossless-agent-skills

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for generate-consistent-og-images

README.md
[![agentmods](https://agentmods.dev/badge/skills/lossless-group/lossless-agent-skills/generate-consistent-og-images/github.svg)](https://agentmods.dev/skills/lossless-group/lossless-agent-skills/generate-consistent-og-images)
Your own site
<a href="https://agentmods.dev/skills/lossless-group/lossless-agent-skills/generate-consistent-og-images"><img src="https://agentmods.dev/badge/skills/lossless-group/lossless-agent-skills/generate-consistent-og-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.

agentmods 80×15 button for generate-consistent-og-images

Your own site · 80×15
<a href="https://agentmods.dev/skills/lossless-group/lossless-agent-skills/generate-consistent-og-images"><img src="https://agentmods.dev/badge/skills/lossless-group/lossless-agent-skills/generate-consistent-og-images.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 432 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 8,385 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00432 $0.08385
Opus 5 $0.00216 $0.04193
Sonnet 5 $0.00086 $0.01677
Haiku 4.5 $0.00043 $0.00839

Measured 11d ago against content hash 3ce2688c8390, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade B, and why

generate-consistent-og-images scanned grade B with 2 findings 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 11d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (templates/ideogram-request.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

# ~/.secrets (chmod 600)

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

No MCP server is required. Ideogram's v3 endpoint accepts a direct HTTPS request with `Api-Key` header. Bash + `curl` is enough. If a future MCP wrapper is preferred, register it the same way the firecrawl/tavily wrapper
generate-consistent-og-images/SKILL.md · 382 lines

How it starts

The opening of the file, as written. The whole thing — 382 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Generate Consistent OG Images

Two techniques, either one valid: generate a wordless illustration with Ideogram (two variables per request — aspect ratio and a one-sentence prompt; everything else locked at the DESIGN.md level), or rasterize the project's own hero/specimen component with qlmanage+sips. Illustration quality is stochastic — some runs land, some don't — while a hero rasterization is deterministic and on-message whenever the hero already says the thing. Pick per project, or per format, and let both coexist.

When to use this skill

  • A site, splash page, plugin page, or fundraise deck needs an OG image generated for the first time
  • An existing OG image needs to be refreshed (rebrand, stale visual, new feature launch)
  • A new format is needed alongside an existing one (e.g. you have banner but now need banner_tall for WhatsApp / iMessage)
  • The user asks for "an image for this page" or "an image for this component" that may or may not follow the OG canon
  • Scaffolding a new Astro Knots site that doesn't yet have a DESIGN.md at all
  • "The link unfurl looks wrong" — the OG image is missing, off-brand, or in the wrong format

Philosophy (Lossless OG preference)

The Lossless Group optimizes share imagery for mobile chat previews — iMessage and WhatsApp first — and treats LinkedIn / Twitter / Bluesky / Slack as concentric rings around that target.

In practice this means:

  • Tall aspect ratios get production attention, not just landscape banners. 3x4 (WhatsApp / iMessage default) and 2x3 (extra-tall) variants are first-class assets, not afterthoughts.
  • Subject lives in the lower portion of the frame by default so the top half stays clean for the receiver's preview crop (chat platforms aggressively top-crop tall images).
  • The 1200×630 OG-canonical banner is one format among several, not the only target. Other social-specific banners (LinkedIn portrait, Twitter Card) are still produced when needed, but never at the expense of a polished tall variant.

Read the full file on GitHub · 382 lines

Files

What ships with it

2 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.

Changes

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.

  1. 11d ago First seen · 382 lines · 432 tokens per session scan B 3ce2688c8390

Subscribe to this mod's changes

generate-consistent-og-images is a skill published in the GitHub repository lossless-group/lossless-agent-skills (4 stars, last pushed 18d ago), licensed MIT. It adds 432 tokens to every session and 8,385 once invoked, about $0.0022 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.