2.6k
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
3 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill windsor-ai- _meta.json292 B
- SKILL.md9.3 KB
Overview
This skill connects to the Windsor.ai MCP to provide natural language access to 325+ data sources, including Facebook Ads, GA4, HubSpot, Shopify, and more. It lets you query, compare, and report on ad, analytics, CRM, and e-commerce data through a single interface. Use it to get fast insights, cross-channel breakdowns, and ready-to-share performance reports.
How this skill works
The skill uses the Windsor.ai MCP to translate plain-English questions into structured queries against your connected data sources. It inspects available connectors, fields, and date coverage in your Windsor.ai account and returns metrics such as spend, conversions, ROAS, revenue, and segment breakdowns. Reports and summaries are assembled from the returned data for quick decision making.
When to use it
- Compare ad spend, ROAS, or CPC across multiple platforms (Facebook, Google, TikTok, etc.).
- Analyze e-commerce performance by product, category, or traffic source (Shopify required).
- Inspect CRM and pipeline metrics like leads, deal size, and conversion rates (HubSpot/Salesforce required).
- Generate weekly, monthly, or client-ready performance reports covering all connected channels.
- Detect anomalies or sudden performance shifts in campaigns over a chosen date range.
Best practices
- Always specify the data source(s), metric(s), time range, and any segmentation for precise results.
- Verify WINDSOR_API_KEY is exported in your shell so the MCP connection can authenticate.
- Use cross-channel queries to reveal attribution and blended ROAS rather than single-source views.
- Start with discovery questions (connected sources, available fields, date coverage) before deep queries.
- Limit date ranges for large accounts to keep queries performant; expand incrementally as needed.
Example use cases
- Compare last 30 days spend and ROAS across Facebook Ads, Google Ads, and TikTok Ads.
- Generate a monthly performance report with executive summary, channel breakdown, and recommendations.
- Find the top 5 campaigns by ROAS last month and the bottom 5 by spend efficiency.
- Break down Google Ads performance by campaign for a specific month including impressions and CPA.
- Show revenue from paid vs. organic traffic for this month (Shopify or GA4 connected).
FAQ
Obtain your WINDSOR_API_KEY from your Windsor.ai account, add it to ~/.clawdbot/.env, and export it in your shell. Then add the windsor-ai MCP entry to config/mcporter.json and verify with npx mcporter list.
Which data sources can I query?
Any source you have connected in Windsor.ai is queryable. Windsor supports 325+ platforms including Facebook Ads, GA4, HubSpot, Shopify, TikTok Ads, and Salesforce.
Why am I seeing authentication errors?
Ensure WINDSOR_API_KEY is set and exported in your active shell session. Restart mcporter after updating environment variables and confirm the key matches your Windsor.ai dashboard.