AI marketing agent and AI ad generator for your agent. Research winning ads, generate on-brand image and video ads, publish, schedule and run ad campaigns. No markup on models.
Hermoso: the marketing MCP
Hermoso is an AI marketing agent and AI ad generator an agent operates end to end: research the ads already winning in any market, create on-brand image and video ads (including UGC-style talking-head spots with native speech), publish or schedule to the user's own channels, and build and read ad campaigns on twelve ad platforms. One MCP server, and every area works on its own: nothing requires another step first. No markup on models.
Connect (once)
- The user creates a free account at https://app.hermoso.ai and makes an agent key on the MCP & CLI tab, under Terminal & API keys
(it starts with
hmk_; direct link: https://app.hermoso.ai/?tab=api-keys, and Settings → API keys goes there too). Free accounts start with credits; renders are quoted before they run. - Add Hermoso as an MCP server in OpenClaw (Settings → MCP → Add server), either transport:
- Streamable HTTP: URL
https://app.hermoso.ai/mcp, headerAuthorization: Bearer hmk_… - Stdio: command
npx, args-y hermoso mcp, environmentHERMOSO_TOKEN=hmk_…
- Streamable HTTP: URL
- Verify it answers:
openclaw mcp doctor hermoso --probe. Then callhermoso_capabilities(free) once; it returns the live model catalog with exact credit costs, which is what to quote before any render.
The tools, by job
- Research:
find_competitors,pull_competitor_ads,research_ads,search_meta_ads,search_google_ads,search_linkedin_ads,search_tiktok,search_instagram,search_youtube,mine_angles,analyze_video. Nearly free. Report the strongest hooks and angles and quote the real headlines. - Create:
draft_brand(onboard a brand from its website once) →plan_ad→render_ad(the Studio pipeline: brand, product photo, storyboard, native audio, read-back of the delivered file).generate_image/generate_videofor raw model runs,make_template_adfor native formats,make_explainerfor narrated explainers. Video renders take 2 to 6 minutes:render_adreturns a job id. Keep callingget_jobuntil it reports done, and never describe a video before that. - Publish & schedule:
post_to_meta(Facebook + Instagram + Threads),post_to_x,post_to_linkedin,post_to_tiktok,post_to_youtube,post_to_pinterest,post_to_reddit,post_to_bluesky,post_to_telegram,post_to_google_business;schedule_post,list_scheduled,reschedule_post,cancel_scheduled.upload_fileturns any local or external file into a URL these accept. Channels are connected by the user in the app (OAuth needs a browser);list_connectorssays which are live. - Ads management:
create_meta_campaign/_adset/_ad,create_google_ads_campaign/_ad_group/_ad, and the TikTok, LinkedIn, Pinterest, Reddit, Microsoft, X and ChatGPT equivalents;meta_insights,google_ads_reportand the per-platform reports. Everything is created paused and read back from the platform before it is described. Only theset_*_ads_statustools withconfirm:truecan make money move; never passconfirmunless the user asked to go live.
Shell shortcuts (same account, via exec)
npx -y hermoso auth login --token hmk_…, then: hermoso competitors <domain>, hermoso ads pull --company "<name>",
hermoso research "<request>", hermoso brand draft --domain <domain>, hermoso create --brand <name> --product "<product>" --format video,
hermoso generate image --prompt "…", hermoso jobs get <id> --wait, hermoso credits, hermoso capabilities. Add --json for machine output.
Rules the agent should keep
- Quote credits (from
hermoso_capabilities) before a render or a campaign build; the user pays per render, not per seat. - A render is done when its URL comes back, not before.
- If a tool answers "not connected", ask the user to connect that channel in the app; the agent cannot complete OAuth.
- Nothing here spends ad budget without the explicit confirm flag.
Related skills
- Downloads
- 527All time
- Security audit
- Pass
- Last updated
- 3w ago
- Current version
- v1.1.0
- License
- MIT-0
