Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/HasData/hasdata-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/hasdata/hasdata-mcp/hasdata-umbrella)<a href="https://agentmods.dev/rules/hasdata/hasdata-mcp/hasdata-umbrella"><img src="https://agentmods.dev/badge/rules/hasdata/hasdata-mcp/hasdata-umbrella/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/hasdata/hasdata-mcp/hasdata-umbrella"><img src="https://agentmods.dev/badge/rules/hasdata/hasdata-mcp/hasdata-umbrella.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00019 | $0.00471 |
| Opus 5.5 | $0.00008 | $0.00188 |
| Sonnet 5.5 | $0.00004 | $0.00094 |
| Haiku 4.5 | $0.00002 | $0.00047 |
Grade A, and why
hasdata-umbrella scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 34 lines — stays where its author put it; the contents beside it link to each section on GitHub.
HasData through one endpoint
https://mcp.hasdata.com/mcp exposes every HasData tool at once, 63 of them, across search, e-commerce, maps, travel, real estate, jobs and social. Authentication is an x-api-key header or OAuth 2.1.
Narrow the surface before you connect
?apis= filters the endpoint to the services you actually need. ?apis=amazon returns five tools instead of sixty-three, and several services combine either comma-separated or by repeating the parameter, so ?apis=amazon,walmart and ?apis=amazon&apis=walmart both return eight.
An unknown service name is an error rather than an empty list, so a typo fails loudly.
Filtering is worth doing. A model choosing among sixty-three similarly shaped tools picks wrong more often than one choosing among five, and every tool description sits in the context window whether or not it gets called.
Tool names say where a tool lives
Names follow hasdata_<service>_<group>_<method>, as in hasdata_amazon_search_getSearchResults or hasdata_google_travel_hotels_getGoogleHotels. The service segment is what ?apis= takes, and it does not always match the product name: Google Search is google_serp, flights are google_travel_flights and hotels are google_travel_hotels.
What is the same everywhere
Every tool returns structured JSON rather than HTML, and every successful call spends credits from the connected account. A call that fails on validation is not billed.
Each service also ships as a standalone server with the same tools behind a shorter URL, which is the better choice when a client only ever needs one site.
Read the per-service rule when you work with a particular site. The traps are specific: prices are strings on Shopify, views are numbers in YouTube search and strings in the video call, and a blocked page on Web Scraping is a successful call with an empty body.
Prompts and resources
Ready-made prompts cover the usual workflows. Offer one of those before composing a tool call by hand.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago Changed · +5 lines 9f1b93b97733
- 21d ago First seen · 29 lines · 19 tokens per session scan A 84d679afc2bc
hasdata-umbrella is a cursor rule published in the GitHub repository HasData/hasdata-mcp (20 stars, last pushed 3d ago), licensed MIT. It adds 19 tokens to every session and 471 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-24.
Other cursor rules, from other repositories
behavior-mcp
This project uses behavior-mcp with project slug "state-memory-mcp" to execute deterministic behavior trees at 60Hz in browser runtimes with reactive triggers and safety guardrails.
browser-controller
Instructions for using Browser Controller to control the user's actual browser.
vision-memory-mcp
This project utilizes vision-memory-mcp to cache visual states, record layout transitions, provide element grounding, and avoid repetitive LLM vision calls.
hasdata-youtube
How to pull YouTube search, video metadata, channels and transcripts through the HasData MCP server.
hasdata-google-maps
How to pull Google Maps places, reviews, photos and posts through the HasData MCP server.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.