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 agentmods add skills/bendrucker/claude/imagenpx skills add bendrucker/claude --skill imagegit clone --depth 1 https://github.com/bendrucker/claudeWhat 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 | $0.00059 | $0.01125 |
| Opus 5 | $0.00030 | $0.00562 |
| Sonnet 5 | $0.00012 | $0.00225 |
| Haiku 4.5 | $0.00006 | $0.00112 |
Grade A, and why
image 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 2d 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.
1. Resolve the image. Use the user's path directly. If they name a meme format instead of a path, check the template library first (below). Otherwise WebSearch for the image and download with `curl -L -o tmp/<name>.jpg < How it starts
The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Meme Image
Overlay meme text on an image with a deterministic renderer. You judge layout, placement, and legibility; the script guarantees the classic look (Impact stroke, subtitle yellow, caption bars).
Workflow
-
Resolve the image. Use the user's path directly. If they name a meme format instead of a path, check the template library first (below). Otherwise WebSearch for the image and download with
curl -L -o tmp/<name>.jpg <url>. Never pass a URL to the renderer. -
Read the image. Note the subjects, where faces and action are, contrast, and aspect ratio.
-
Choose a mode:
- Top/bottom macro text:
--top/--bottom - TV-subtitle quote (yellow, bottom center):
--subtitle - White bar above or below the image:
--caption(optionally--caption-position top|bottom, default top) - Multi-label formats or custom placement: write a JSON spec file. Read references/spec.md first.
- Top/bottom macro text:
-
Pick the output filename. It is part of the joke: witty and meme-relevant, kebab or snake case,
.png. Nevermeme.pngor another generic name. If the user supplies a filename, use theirs. -
Render, then Read the output PNG. Check legibility, that text does not cover key subjects, line-break placement, and any fit warnings on stderr. Adjust and re-render until it reads well.
-
Deliver: copy the file reference to the clipboard so a paste keeps the filename:
osascript -e 'set the clipboard to POSIX file "<absolute output path>"'Report the path. If the clipboard copy is declined or fails, reveal the file instead:
open -R <path>.
Template Library
${CLAUDE_PLUGIN_DATA}/templates/ holds the user's meme templates, synced outside git. Never copy its images into the repo. Each image may have a sidecar spec of the same basename (drake.jpg + drake.json): a ready-made layout whose text values are <slot descriptions> and whose description says how the format works.
- User names a format:
lsthe library and match by filename. With a sidecar, copy it totmp/, replace each slot with the actual joke, and render with--spec. Panel counts matter: fill every slot the joke needs and drop boxes the format leaves empty. - No sidecar (new template): Read the image, trace regions, render, iterate. Once the layout looks right, save it back as a sidecar with
<slot description>placeholders so the next use skips the tracing.
What ships with it
11 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.
- .gitignore 31 B
- references/spec.md 3.8 KB
- scripts/__snapshots__/spec.test.ts.snap 1.5 KB
- scripts/layout.test.ts 3.8 KB runs code
- scripts/layout.ts 3.4 KB runs code
- scripts/presets.ts 2.1 KB runs code
- scripts/render.test.ts 3.6 KB runs code
- scripts/render.ts 10 KB runs code
- scripts/setup.ts 495 B runs code
- scripts/spec.test.ts 3.1 KB runs code
- scripts/spec.ts 3.1 KB runs code
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.
- 2d ago First seen · 68 lines · 59 tokens per session scan A bfcbd27d6189
image is a skill published in the GitHub repository bendrucker/claude (16 stars, last pushed 2d ago), licensed MIT. It adds 59 tokens to every session and 1,125 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
media-ingest
Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.
diagnostic-stem-delivery
Audio production with diagnostic analysis, timecode parsing from documents, and verified export workflow.
vox-director
Turn ONE topic into a finished Vox-style paper-collage explainer / ad video, end to end on the Atlas Cloud API + local ffmpeg — script, collage keyframes, motion, voice-over, music, captions, all automated. Use this whenever the user wants a "Vox style" video, a paper/torn-paper collage animation, a "motion collage"…
seedance-vocab-ja
This skill should be used when the user asks for Japanese Seedance 2.0 prompt wording, Japanese cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms into Japanese.
model-compatibility
Model family compatibility matrix covering loaders, resolutions, samplers, CFG, VAE, ControlNet, and LoRA compatibility for SD 1.5, SDXL, Flux, SD3, and video models.
comfyui-topology-viz
Turn a network topology into one stylized, AI-generated still image via a self-hosted ComfyUI instance — reuses the same topology model as threejs-network-viz (any of 8 topology-source integrations, or a freeform description). Use when the operator asks for a stylized, flashy, or AI-generated…