Raven is an open-source agent harness for running long-term AI work with terminal execution, tracing, memory, skills, evaluation, and reusable workflows. People use the current release to operate and improve persistent AI workflows, while its described future direction is a multi-agent system that combines specialized harnesses.
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 EverMind-AI/Raven --skill create-marketing-graphicsgit clone --depth 1 https://github.com/EverMind-AI/RavenWrote 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/evermind-ai/raven/create-marketing-graphics)<a href="https://agentmods.dev/skills/evermind-ai/raven/create-marketing-graphics"><img src="https://agentmods.dev/badge/skills/evermind-ai/raven/create-marketing-graphics/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/evermind-ai/raven/create-marketing-graphics"><img src="https://agentmods.dev/badge/skills/evermind-ai/raven/create-marketing-graphics.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 | $0.00378 | $0.10026 |
| Opus 5.5 | $0.00151 | $0.04010 |
| Sonnet 5.5 | $0.00076 | $0.02005 |
| Haiku 4.5 | $0.00038 | $0.01003 |
Grade A, and why
create-marketing-graphics 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 — 384 lines — stays where its author put it; the contents beside it link to each section on GitHub.
创建营销传播图形(海报与套系)
核心目标
让没读过任务的人在三秒内读懂这张海报在说什么、五秒内相信它已经贴在街上或发在账号上。交付两样东西: 成品平面稿(每个触点、每个尺寸一张最终像素,像真实发布的海报,不打水印、不标"示意"、不用占位)和 母版文件夹(HTML/CSS 画板母版、内容插槽、素材层、色与字规范、一条命令重出全部成品)。成品给人看效果 做决定,母版让人能改内容接着出。我们做的是示例与方向,不是印刷厂文件;但示例必须像真的一样。
成品是最好的参考效果图:让看的人感到这件事能做到的上限,而不是把人的感知拉到"文档级"的下限。任何"怕被误认成
真实照片"的顾虑都用 README.md 与 POSTER-BRIEF.md 里的来源说明解决,从不用降级画面解决。
遵循五个原则:
- 分层版面是基底。 任何海报都是五层:画板 → 背景层 → 主视觉层 → 图形元素层 → 文字与品牌层。
基底是固定尺寸的 HTML/CSS 画板,不是一张生成图;版面与全部活字只在 HTML 画板里排,SVG 只做 logo 与符号的资产格式,
用 SVG 坐标排版面即基底失败。每张成品的主视觉或场景级背景必须是
image_generate产出的位图; 纯排版方向也不例外,它的背景层是有构图、光源与景深的场景级生成图。无缝材质、纸纹、噪点不算生成层。 - 一个焦点、一个主视觉、一种做法。 主视觉只有一个,只用一种做法(照片 / 抠图 / 插画 / 3D 渲染,均为生成图 或用户真图;或纯排版),不叠加第二种;标题是唯一的第一眼。手绘几何、线描图标、"抽象形态"不是主视觉做法。
- 活字不画字。 标题、副标、日期地点、价格、CTA、logo 全部是真实字体的活字或矢量;生图只出背景与 主视觉素材,prompt 里明确不要文字。
- 系列只变内容不变容器。 画板、背景做法、图形元素、标题位置、信息层位置、logo 位置是不变项; 变的只有主视觉内容与文案。
- 像真的一样。 真名称、真日期、真地点、真价格、真 logo;场景可信、拼写正确;缩小到拇指大仍能认出。用户没给的
事实用可信替身做到最终效果,并在
README.md的《替换清单》逐项列出;成品画面上永远不出现"待替换 / 待填 / 预留"。
本 Skill 只拥有海报与营销图形的领域判断。流程与证据服从共享 $visual-artifact-design。长期品牌系统转交
品牌领域并继承其批准资产;编辑出版物与学术海报转交编辑领域;独立插画与数据图表转交对应领域;只有最终
消费者是网页时才组合 $build-polished-visual-frontends,HTML 画板不把海报变成网站。
最终消费者、目标媒介和观看环境必须写入交付合同,并以最终像素或等价终态验收;字体、图像、商标和生成资产记录来源、许可与替换边界。
什么算好,七条判据(后文以代号引用):
- P1 唯一焦点:标题或主视觉之一是第一眼,其余都退后;缩到 200 px 高仍先看到它。
- P2 三级层级:钩子(标题 / 主视觉)→ 支撑(副标、日期地点)→ 细则(票价、二维码、赞助、法务), 信息层默认集中在底部约 1/5;参考或阅读顺序支持其他位置时,可整体调整并在最小系统合同中记录依据, 但日期、地点与行动入口必须容易定位,不能无规则散落。
- P3 一个主视觉一种做法:满版照片、抠图主体、插画、3D 渲染、纯排版只选一种;没有第二个抢眼视觉;主视觉不是手绘几何。
- P4 活字正确:所有文字逐字正确、字体是所选字体、字号符合观看距离;画面里没有生成的伪文字。
- P5 系列同容器:并排缩略像一套;不变项逐张 diff 为零;变量只在主视觉与文案。
- P6 尺寸即规格:每个触点用它真实的像素或毫米尺寸出图,安全区内有标题与 CTA,不是同一张图缩放。
- P7 像真的:没读过任务的人把它当成真实发布的海报;真 logo、真信息、可信场景,没有模板占位感。
0. 入口 gate:先判断这是不是本 Skill 的任务
用户不会说"我要海报",会说"做个图 / 做个宣传 / 做个封面"。读到本 Skill 不等于任务属于本 Skill。开始任何工作前对照 下面三条,三条同时成立才继续;任何一条不成立,在回复里写一行"不属 create-marketing-graphics,改按 <Skill 名>" 并转到对应 Skill,不许先做再改,也不许自由归类:
- 输出是定尺寸单帧静态平面文件:像素或毫米定死,交付 PNG/JPG/PDF;不可滚动、不可交互、不分页。
- 一帧承载全部信息:读者一次扫读完,不逐页读、不逐段读。
- 用途是传播:宣传、告知、售卖、招募、邀请、展示。
What ships with it
23 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.
- agents/openai.yaml 492 B
- references/deliverable-skeleton.md 4.1 KB
- references/implementation-routing.md 12 KB
- references/patterns.md 9.0 KB
- references/poster-examples/anatomy.jpg 675 KB
- references/poster-examples/brand-boss.jpg 304 KB
- references/poster-examples/brand-meituan.jpg 563 KB
- references/poster-examples/domain-brand.jpg 543 KB
- references/poster-examples/domain-event.jpg 498 KB
- references/poster-examples/domain-promo.jpg 494 KB
- references/poster-examples/domain-travel.jpg 549 KB
- references/poster-examples/event-goldenhorse.jpg 686 KB
- references/poster-examples/event-wuzhen.jpg 613 KB
- references/poster-examples/INDEX.md 3.6 KB
- references/poster-examples/manifest.json 5.4 KB
- references/poster-examples/multisize-converging.jpg 519 KB
- references/poster-examples/promo-luckin-duolingo.jpg 431 KB
- references/poster-examples/promo-luckin-linedog.jpg 469 KB
- references/poster-examples/promo-tmall.jpg 396 KB
- references/poster-examples/series-fis.jpg 775 KB
- references/poster-examples/series-hippocampus.jpg 528 KB
- references/poster-examples/series-taobao.jpg 459 KB
- references/tool-profiles.md 9.8 KB
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 First seen · 384 lines · 378 tokens per session scan A f7cfeb632d81
create-marketing-graphics is a skill published in the GitHub repository EverMind-AI/Raven (5,463 stars, last pushed today), licensed Apache-2.0. It adds 378 tokens to every session and 10,026 once invoked, about $0.0015 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 2026-09-12.
Other skills, from other repositories
dcf-model
Build discounted cash flow valuation workbooks in Excel.
comps-analysis
Build comparable-company valuation workbooks in Excel.
anatomy
Explain an idea as an interactive isometric machine figure.
mcp-oauth-remote-gateway
Manual OAuth for remote MCP servers on headless gateways.
p5js
Use when users request: p5.js sketches, creative coding, generative art, interactive visualizations, canvas animations, browser-based visual art, data viz, shader effects, or any p5.js project.
honcho
Configure and troubleshoot Honcho memory for Hermes.