Add TensorFeed (Specialised Dataset, Open auth)

[RipperMercs]

Server

Name: TensorFeed
Category: Specialised Dataset
URL: https://tensorfeed.ai/api/mcp
Auth: Open
Maintainer: TensorFeed

What it does

A single remote MCP server providing aggregated, attributed AI ecosystem data across multiple domains via 17 free tools. Built for the case where an agent needs cross-domain context (security, news, model pricing, regulatory data) without wiring N parallel API integrations.

Tools include: get_cve_record, get_kev_catalog, get_epss_score, query_osv_vulnerabilities (security advisories from MITRE/CISA/FIRST.org/OSV); get_news_articles (100+ AI sources); get_models (cross-lab pricing); get_status_summary (OpenAI/Anthropic/Google/etc.); search_sec_edgar (SEC filings); query_fda_drug_recalls, query_fda_drug_labels, query_fda_device_events, query_fda_food_recalls, query_fda_drug_events (FDA regulatory data); plus US energy/economy indicators via FRED, EIA, BLS.

Why it meets the quality criteria

• Official Support: Maintained directly by Pizza Robot Studios (the maintainers of TensorFeed.ai).
• Production Ready: Live since early 2026, daily 100K+ KV operations, daily cron-driven data refreshes.
• Active Maintenance: 50+ commits/month at https://github.com/RipperMercs/tensorfeed.
• Security: Open auth on free tools (no PII handled). All paid endpoints use a separate x402 V2 payment surface (Coinbase exact scheme on Base mainnet) and are not exposed via the MCP surface. Worker enforces request validation, OFAC screening, rate limits, and chaos/circuit-breaker subsystems.
• Reliability: Cloudflare Workers + Pages, multi-cron data freshness, AFTA-certified self-attested at https://tensorfeed.ai/api/afta-certify/check?domain=tensorfeed.ai.
• Community: Listed in the canonical MCP registry as ai.tensorfeed/mcp-server, included in the Anthropic financial-services and life-sciences marketplaces, on Hugging Face as tensorfeed/ai-ecosystem-daily, and shipped with skill packages in anthropics/skills (PR #1114) and openai/skills (PR #405).

Production usage examples

• Cross-database CVE verification (a single agent loop calls get_cve_record + get_kev_catalog + get_epss_score and returns a confirmed_by list): see the Anthropic claude-cookbook example and the OpenAI cookbook example.
• Hosted MCP integration: see the openai-agents-python example and the claude-agent-sdk-python example.

Notes

• Underlying data is mostly US Government public domain (SEC, BLS, FRED, MITRE, CISA, EIA) or CC0 (openFDA), with FIRST.org EPSS free for any use; commercial redistribution permitted with attribution preserved on every response.
• Open auth is appropriate here because the free tools serve only public, government-or-commons-licensed reference data with no per-user state.