Open-Source Cold Email & Warmup Platform
Warmbly 是開源的冷郵件平台:溫暖您的收件箱、發送活動、並對每個回覆進行分流。可在我們的雲端運行或自行託管整個系統。
AI 翻譯
ContentHero Skills: the context-plus-execution layer for your agent. Generate media, run the content pipeline (ground in your outliers and brand voice, draft on-brand, publish and schedule), and read your brand context. Your LLM stays the brain.
翻譯準備中
取自儲存庫 README
正在載入完整 README。 以下為已儲存的摘錄。
AI agent skills for creating and publishing on-brand content. ContentHero is the context-plus-execution layer. Your own LLM is the brain.
Works with Claude Code, Cursor, Codex, OpenClaw, and other AI coding agents.
Paste this into your agent. It clones the repo to the right path, wires a transport (MCP or CLI), grounds itself on your brand kit, and offers a free smoke test.
Prefer to install manually? See INSTALL.md. The short version:檢視完整 README ↗Create, edit, schedule, and publish content from your AI assistant. Generate images, video, and voiceovers, plan your content calendar, and draft on-brand copy, from hooks and scripts to captions and ads, grounded in what already performs for you.
One plugin: the contenthero skill plus the ContentHero connector. It works in Claude, ChatGPT,
Gemini, Codex, Cursor, VS Code, and other AI assistants that read skills.
"Find my top outliers, ground a Reel caption in my brand voice, draft it for my approval, then schedule it."
→ reads your inspiration outliers → extracts the patterns that perform for you → pulls your brand kit
voice and your recent posts → drafts an on-brand caption in your own voice → you approve → produces
the media → schedules the post
Pick your AI at contenthero.ai/skills for step-by-step instructions and the download. The short version is in INSTALL.md:
contenthero-plugin.zip and upload it under Customize, Plugins.contenthero.zip and upload it as a skill, then connect ContentHero.claude plugin marketplace add contenthero-ai/skills, then claude plugin install contenthero@contenthero.ContentHero pairs your LLM with your context. Your LLM is already a strong writer. What it lacks is your brand voice, your proven patterns, and a way to publish. ContentHero supplies all three.
When a draft is needed, the skill grounds your LLM in your real context and your proven patterns, your LLM writes it in your voice, and you approve before anything goes live.
One skill, /contenthero, covering the whole product. It routes your request to the right
workflow rather than making you pick a skill first.
| Workflow | What it covers |
|---|---|
| Generate | Image, video, audio, reference boards, lip-sync, upscaling. Cost preflight before every spend, output-id chaining so you never re-upload. |
| Plan and publish | Spaces, stages and cards, one post per platform, scheduling, and publishing to your connected accounts. |
| Ground and draft | Your brand kit and its searchable knowledge base. Your LLM writes in your voice; you approve before anything goes live. |
| Research | The accounts you track and their posts ranked by outlier score, which measures a post against its own creator's baseline rather than the platform's. |
| Media library | Upload, import, search, folders, and archiving that is always reversible. |
| Editor | Projects, timelines, canvases, renders of your work, and exports. |
It remembers your context in a small workspace file, .contenthero/context.md, caching your
active brand kit and identity ids so they are not re-discovered every run.
It was three separate skills until September 2026. One covers the same ground with a single install, and the real flows crossed all three anyway.
Two transports, detected in priority order:
| Priority | Transport | Trigger | Billing |
|---|---|---|---|
| 1 | MCP (OAuth) | ContentHero MCP tools visible | Your ContentHero account credits |
| 2 | CLI (API key) | contenthero auth status exits 0 or CONTENTHERO_API_KEY set | Your ContentHero account credits |
Never paste an API key into an agent chat. Use contenthero login (browser-assisted), the CONTENTHERO_API_KEY environment variable, or MCP OAuth. See INSTALL.md for details.
| Prompt | What happens |
|---|---|
| "Generate a product image with nano-banana-2, then animate it into a 5-second clip." | generate → chain the output id straight into a video, no re-upload |
| "Find my top outliers this month and tell me the hook patterns that work for my audience." | reads inspiration outliers, extracts pattern primitives |
| "Draft an on-brand caption for this reel using my voice and my best-performing posts." | grounds your LLM in brand voice plus your top posts, drafts for approval |
| "Turn this idea into an on-brand post and schedule it to my connected accounts." | full pipeline: ground → draft → produce → assemble → schedule |
| "Read my brand kit and summarize my tone and banned words." | brand context read |
@contenthero/sdk, @contenthero/mcp, @contenthero/cli)MIT