2026年6月30日

Agentbrush

Your coding agent's missing tool: image generation
AI WritingContent GenerationResearchEmail WritingSummarizationRewritingAcademic Research浏览器扩展

概览
Agentbrush 是什么?

AgentBrush is an MCP server that lets AI coding agents automate image generation inside their workflows: the agent calls and chains the tools (generate, remove-background, save-as-reference) without leaving the editor. Feed it your brand once (colors, fonts, reference images) and every product shot, app icon, OG card, game sprite or mascot lands on-brand and drops straight into your repo. Includes style presets, reference-image

Your coding agent's missing tool: image generation

核心功能
Inline image generation inside your AI coding agent (Claude Code, Cursor, Windsurf, any MCP client)
On-brand by default: feed it your colors, fonts and reference images once and every generation matches
Reference-image locking so a product or character stays identical across scenes
Automatic tool chaining: the agent runs generate, remove-background and save-as-reference itself as a step in the task
Free local background removal for clean transparent PNGs (0 tokens)
Mask editor for precise region edits without re-rolling the whole image
Legible in-image text, including non-Latin scripts
Style presets (realistic, custom and more)
Files save straight into your repo, no tab-switching and no API wiring
热门使用场景
  • Generate full product photography and ad campaigns without a studio
  • Create game assets and pixel-art sprites for Unity, Godot or Phaser
  • Produce on-brand marketing visuals: hero images, OG cards, app icons, mascots, social creatives
  • Generate OG and social preview images so shared links do not render as blank grey boxes
  • Make logos and app icons in-flow while you build
  • Batch-generate consistent images, one per product or SKU
  • Create multilingual marketing creatives with legible headlines
  • Produce transparent PNG cut-outs for marketplace or store listings
如何使用
  • AgentBrush runs as an MCP server that your AI coding agent talks to. Setup takes about two minutes. 1. Create an account at https://agentbrush.dev and choose a plan. 2. Install the server: npm install -g @agentbrush/mcp-server 3. Authenticate: run "agentbrush-mcp init" (it opens your browser
  • click Approve). 4. Connect it to your agent. - Claude Code: claude mcp add agentbrush --scope user -- agentbrush-mcp - Cursor or any other MCP client: add it to your mcp.json: { "mcpServers": { "agentbrush": { "command": "agentbrush-mcp" } } } 5. Now just ask your agent in plain language
  • for example "generate a hero image and an OG card for this landing page." It calls AgentBrush and saves the files straight into your repo.
定价
Agentbrush 采用 Freemium 定价模式,价格和功能可能会随时间变化。
Starter
$6.99 / month
100 tokens per month at low and medium quality. The affordable entry plan for real work, roughly 20 medium or 100 low-quality images a month. Hard usage cap, so no surprise charges.
Pro
$14.99 / month
600 tokens per month and unlocks high quality. The best value and the most popular plan, with a much lower cost per image than Starter. Hard usage cap.
Power
$29.99 / month
1300 tokens per month at all quality levels, and the only plan with pay-as-you-go overage at $0.04 per token so it never cuts off mid-run. Built for heavy users.
优惠 / 优惠码
暂无优惠码。
大家喜欢它的原因
暂时还没有评价标签。
还有哪些可以改进
暂时还没有评价标签。
常见问题

产品时间线
待核实
Agentbrush 评价
社区已经提交了 0 条关于 Agentbrush 的评价,分享他们喜欢什么、哪些地方还可以改进,以及更多实际体验。
基于 0 条评价
Ease of Use
Output Quality
Reliability
Value for Money
Customization
评价 Agentbrush?
优点 (0)
暂无标签。
缺点 (0)
暂无标签。
还没有评价。
MCP Server
验证信息
工具状态
待核实
定价已核验
待核实
创始人已认领
否 / 待核实
来源
官网 / 社区提交
相关标签
AI WritingContent GenerationResearchEmail WritingSummarizationRewritingAcademic Research浏览器扩展Freemium
你是这个工具的官方团队吗?
认领这个资料页后,你可以更新产品信息、定价和官方回复。