synthetic-sciences/openscience is an AI workbench that carries out scientific research by reading papers, forming hypotheses, writing and running code, conducting experiments, analyzing results, and preparing reports. Researchers use it for work in machine learning, biology, physics, and chemistry with remote or local models. Catalogue add-ons extend its scientific workflows through skills and instructions.
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 synthetic-sciences/openscience --skill generate-imagegit clone --depth 1 https://github.com/synthetic-sciences/openscienceWrote 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/synthetic-sciences/openscience/generate-image)<a href="https://agentmods.dev/skills/synthetic-sciences/openscience/generate-image"><img src="https://agentmods.dev/badge/skills/synthetic-sciences/openscience/generate-image/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/synthetic-sciences/openscience/generate-image"><img src="https://agentmods.dev/badge/skills/synthetic-sciences/openscience/generate-image.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 22 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00103 | $0.01590 |
| Opus 5.5 | $0.00041 | $0.00636 |
| Sonnet 5.5 | $0.00021 | $0.00318 |
| Haiku 4.5 | $0.00010 | $0.00159 |
Grade A, and why
generate-image 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 1mo 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 — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Generate Image
Generate or edit an image with the native generate_image tool. The tool keeps credentials
in the trusted host, renders through whichever route the person has, and saves the image
into the connected workspace, where read opens it for inspection and a manuscript
includes it directly.
The route
The environment line Image generation: says whether a route exists and which one:
- Ace: the managed account. Renders Nano Banana Pro (
google/gemini-3-pro-image) and settles the cost from the Wallet, like any other Ace request. One reference image per request. - A Gemini key the user connected in Customize → Models: Nano Banana Pro
(
gemini-3-pro-image) on their own account, up to 14 reference images. - An OpenAI key the user connected: GPT Image 2 (
gpt-image-2) on their own account, reference images as an edit.
A personal OpenRouter key is not a route. When the line says image generation is unavailable, say once that it needs Ace or a Gemini or OpenAI key connected in Customize → Models, then continue with the rest of the request. Never ask the user to paste a key into chat, never call the tool to "check", and never draw the image by hand as TikZ, SVG, Graphviz or matplotlib shapes in its place.
Tool contract
Call generate_image with:
prompt(required): what the image shows and how. Prompt quality decides output quality more than the model does. Name, in one sentence each: the subject and how much of it is in frame ("a single pipette tip above a 96-well plate"); the medium and style (scientific illustration, flat vector, watercolour, 3D render, photograph); the lighting and palette ("soft diffuse light, cool blue and white palette"); the composition ("wide shot, subject left of centre, empty space on the right for a title"); and what to avoid ("no text, no labels, no watermark"). Any text that must appear is quoted verbatim. Asking for empty space where a caption or title will go is the single most useful compositional instruction for posters and slides.purpose:illustrationfor a conceptual figure or graphical abstract (the tool prepends publication framing: clean background, restrained palette, no invented text or data);editwhen changing an existing image;schematicbelongs to the schematics skill.output_path: destination in the workspace,.png(any route),.jpgor.webp(Ace and OpenAI). Defaultgenerated-image.png.input_path: an existing image to edit. Omit it entirely for a new image; never pass a directory,.,/dev/nullor a blank canvas.reference_paths: existing images whose style or components the result should follow. Up to 14 on a Gemini or OpenAI key; on Ace pass one image in total (eitherinput_pathor one reference).aspect_ratio:1:1,3:2,2:3,4:3,3:4,16:9,9:16or21:9, from the slot the image fills.image_size:1Kwhile iterating (default),2Kfor anything printed,4Konly for a poster or cover.
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.
- 1mo ago Changed · +48 lines · +36 tokens per session c7a03d15f73d
- 1mo ago First seen · 78 lines · 67 tokens per session scan A 55ec6967055e
generate-image is a skill published in the GitHub repository synthetic-sciences/openscience (4,160 stars, last pushed yesterday), licensed Apache-2.0. It adds 103 tokens to every session and 1,590 once invoked, about $0.0004 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 2026-09-03.
Other skills, from other repositories
system-prompts
Write system prompts, tool docs, and agent definitions. Project tag conventions + RFC 2119 keywords + dense compression. Use when authoring or editing any prompt the model reads.
tool-prompt-optimization
Optimize the description prompts an AI agent reads to learn its built-in tools (the .md files under prompts/tools/). Two halves: (1) measure how much of a prompt is already inferable from the tool's JSON parameter schema + name, to prune redundancy with evidence; (2) house authoring rules for what belongs in a tool…
gno
Search local documents, files, notes, and knowledge bases. Index directories, search with BM25/vector/hybrid, get AI answers with citations. Use when user wants to search files, find documents, query notes, look up information in local folders, index a directory, set up document search, build a knowledge base, needs…
knowledge-model-and-provider-runtime-provider-resilience-admission-and-observability
A codebase guide for handling unreliable AI provider requests across agents and sessions, including retries, circuit breakers, fair queues, stalled streams, cache tracking, and health checks.
journal
A mirror for emotional truth. Use when the user wants to journal, reflect, or "tell me the truth about how I feel." Goes for the feeling beneath the event, names avoidance, and refuses easy answers — not a cheerful prompt list.
knowledge-model-and-provider-runtime-model-catalog-configuration-and-credentials
A codebase guide for defining AI models and providers, selecting endpoints and credentials, configuring fallback channels, and migrating model identifiers.