create-image-fal
Generate or edit an image via any FAL image model (nano-banana edit, gpt-image, flux, ...), ROUTED THROUGH THE fal-proxy so it bills the Ads agent. image_urls…
Evaluate a website, landing page, content, or any online asset through the eyes of pre-built synthetic ICP personas. Loads personas from buyer-persona-generator output, then runs them against target URLs. Supports three modes: structured scorecard, freeform focus group, and
$ npx -y skills add gooseworks-ai/goose-skills --skill icp-website-review --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/icp-website-reviewContext preview
The summary Claude sees to decide when to auto-load this skill.
Evaluate a website, landing page, content, or any online asset through the eyes of pre-built synthetic ICP personas. Loads personas from buyer-persona-generator output, then runs them against target URLs. Supports three modes: structured scorecard, freeform focus group, and
name: icp-website-review description: > Evaluate a website, landing page, content, or any online asset through the eyes of pre-built synthetic ICP personas. Loads personas from buyer-persona-generator output, then runs them against target URLs. Supports three modes: structured scorecard, freeform focus group, and head-to-head competitive comparison. Reusable — run against the same site after changes, or against new content anytime. tags: [research]
Evaluate any online asset through the eyes of pre-built synthetic ICP personas. This is the evaluation engine — it loads personas created by `buyer-persona-generator` and runs them against whatever you point it at.
Personas must exist at `clients/<client>/personas/personas.json`. If they don't, run `buyer-persona-generator` first.
**Structured scorecard (default):**
Review [company]'s website using their ICP personas. Site: [url].
**Freeform focus group:**
Run a focus group on [company]'s new landing page: [url]. Use focus-group mode.
**Head-to-head competitive:**
Compare [company]'s site ([url]) against [competitor] ([url]) through ICP personas.
| Input | Required | Source | |-------|----------|--------| | **Personas** | Yes | `clients/<client>/personas/personas.json` | | **Target URL(s)** | Yes | User provides | | **Mode** | No | `scorecard` (default), `focus-group`, or `head-to-head` | | **Competitor URL(s)** | For head-to-head | User provides | | **Specific pages** | No | Default: homepage + key pages discovered during crawl | | **Scope** | No | `full-site` (default) or specific page URL for a single-page review |
Rigorous, dimension-by-dimension evaluation with numerical scores. Best for:
Each persona reacts naturally — stream of consciousness, emotional reactions, questions they'd have, things that confuse them. No rigid scoring. Best for:
Each persona evaluates your site AND a competitor's site, then picks a winner with reasoning. Best for:
---
Read `clients/<client>/personas/personas.json`. Confirm:
Fetch the key pages each persona would realistically visit:
1. **Homepage** (everyone starts here) 2. **Pricing page** (if public) 3. **Product/features page** (technical buyers) 4. **Solutions/use-cases page** (business buyers) 5. **About page** (trust/credibility check) 6. **Case studies or testimonials** (social proof) 7. **Blog** (1-2 recent posts, for content quality signal) 8. **Documentation or resources** (if relevant for technical buyers)
Use `WebFetch` for each page. Extract main content, headlines, CTAs, and overall structure. Note what's present and what's missing.
Also check external presence:
For **head-to-head mode**, crawl the same pages on the competitor site.
---
For each persona, evaluate on these dimensions:
**A) First Impression (10 seconds)**
**B) Messaging Relevance**
**C) Trust & Credibility**
**D) Clarity & Information Architecture**
**E) Objection Handling**
**F) Overall Verdict**
---
For each persona, produce a natural, first-person reaction. No rigid structure — let the persona talk.
**Prompt each persona with:** > You are [persona name], [title] at [company type]. [Situation summary]. You just landed on this website. React naturally — what do you notice first? What confuses you? What excites you? What makes you skeptical? Would you keep exploring or bounce? If you stayed, where would you go next? What questions do you need answered before you'd take action?
**Guidelines:**
Put your AI agent on the growth team. Research customers and competitors, analyze what is working, create the next campaign, and learn from the result.
Repo: gooseworks-ai/goose-skills
Generate or edit an image via any FAL image model (nano-banana edit, gpt-image, flux, ...), ROUTED THROUGH THE fal-proxy so it bills the Ads agent. image_urls…
Generate a single photoreal or designed image with OpenAI gpt-image via fal.ai. Supports gpt-image-1 (default, fixed sizes — the FAL fallback for Higgsfield's…
Generate an instrumental music bed via ElevenLabs Music, ROUTED THROUGH THE elevenlabs-proxy so it bills the Ads agent. Trims any sparse intro, loudnorm, fades…
Image-to-video (or text-to-video) via any FAL video model (Kling, Seedance, Veo), ROUTED THROUGH THE GooseWorks fal-proxy so the call bills the Ads agent. The…
Generate a voiceover (VO) clip via ElevenLabs text-to-speech, ROUTED THROUGH THE elevenlabs-proxy so it bills the Ads agent. Voice id + script text come from…
Scrape competitor ads from Google Ads by domain. Returns ad creatives, formats, and campaign details. Use for competitive ad research and messaging analysis.