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 mphinance/alpha-skills --skill image-bakeoffgit clone --depth 1 https://github.com/mphinance/alpha-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/mphinance/alpha-skills/image-bakeoff)<a href="https://agentmods.dev/skills/mphinance/alpha-skills/image-bakeoff"><img src="https://agentmods.dev/badge/skills/mphinance/alpha-skills/image-bakeoff/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/mphinance/alpha-skills/image-bakeoff"><img src="https://agentmods.dev/badge/skills/mphinance/alpha-skills/image-bakeoff.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 | $0.00077 | $0.00791 |
| Opus 5.5 | $0.00031 | $0.00316 |
| Sonnet 5.5 | $0.00015 | $0.00158 |
| Haiku 4.5 | $0.00008 | $0.00079 |
Grade A, and why
image-bakeoff scanned grade A with 0 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 9d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
image-bakeoff
Serial runs of N models x M prompts through the openrouter-images skill, saved with real cost and timing, then rendered as a models-by-prompts grid. Built from the 2026-10-08 bakeoff (winner: openai/gpt-image-2, see memory image-model-choice-gpt-image-2).
Workflow
- Pick models. Check live prices first:
cd ~/.claude/skills/openrouter-images/scripts && npx tsx discover.ts <model>(pricing lines per endpoint; token-priced models have no per-image price, so estimate or measure). Also seelist-daily-model-rankingswithmodality=image_outputfor usage. - State the estimated cost to the user before running (models x prompts x $/image). Per the global cost rule, never batch without saying so. Typical: $0.007 to 0.09 per image; 8 models x 2 prompts is about $0.6.
- Write prompts to a JSON file:
{"p1": "text", "p2": "text"}. Mix a text-in-image prompt, a photoreal portrait, a product shot, a stylized illustration. - Run (serial; some models take 60-90s per image, so use a background run for big sets):
Re-running into the samepython3 ~/.claude/skills/image-bakeoff/scripts/bakeoff.py run --models openai/gpt-image-2,recraft/recraft-v4.1-flash --prompts prompts.json --out ~/projects/openrouter/imgbakeoff-YYYY-MM-DD--outaccumulates intoresults.json(same model+prompt replaces). Extra generate.ts flags:--gen-args "--quality high". - Sheet:
Writespython3 ~/.claude/skills/image-bakeoff/scripts/bakeoff.py sheet --results DIR/results.json --models a/b,c/d --name combined --title "..."<name>.htmland<name>.pngnext toresults.json. Filter with--models/--promptsto keep the sheet wide, not tall (models are rows, prompts are columns). Always view the PNG after rendering to confirm nothing is clipped. - Save everything (images, prompts, results.json, sheet) in a project folder, never the scratchpad. Always report cost and time per model.
Gotchas
- Time includes ~1-2s
npxstartup, and results.json stores seconds measured by the script, not by the API. - Some models report no cost (cost shows
n/a), e.g. nano-banana-2.1. - Output extensions vary by model (png/jpg/webp); the script records the real saved path.
- Hand-built manifests may span folders: cell
fileis relative to the manifest's directory, and an optional"tiers": {"model": "CHEAP"}adds a label. - Cells are center-cropped squares in the sheet; open the originals before judging framing (e.g. tall posters get cropped).
- Image price scales with quality/size for token-priced models (gpt-image-2 family); default-quality runs are not representative of
--quality high.
What ships with it
1 file 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.
- 9d ago First seen · 35 lines · 77 tokens per session scan A 719bc62bbfbd
image-bakeoff is a skill published in the GitHub repository mphinance/alpha-skills (28 stars, last pushed 10d ago), licensed MIT. It adds 77 tokens to every session and 791 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-10.
Other skills, from other repositories
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…
next-cache-components-optimizer
Optimize the meaningful UI available immediately from a Next.js route on an initial load (hard navigation) or named client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route and entry point at a time; the shipped test then guards…