Where Mindber data comes from
Two ingest paths feed every product page: a small set of public catalog sources for identity and metadata, and a public engagement signal layer that produces the Mindber Activity Score. Both run on a fixed cadence and write to a versioned audit trail.
Catalog sources
Public AI-tool directories Mindber ingests for product identity, launches, categories, and rankings. Each source links back to its raw profile on the source platform.
- Product HuntActive
- Launches, maker metadata, category tags, votes/comments, awards via Product Hunt GraphQL API. Primary discovery for new products.
- Daily 06:00 UTC (API)
- ToolifyActive
- Official tool listing pages only — not compiled third-party traffic reports. Monthly traffic estimates, ranking signals, regional rankings, pricing snapshots, AI-model and platform metadata.
- Daily 04:00 UTC
- TAAFT (There's An AI For That)Active
- Tool discovery, country, creator socials, Q&A, alternatives, deals.
- Tue 04:00 UTC
- AIbasePaused
- China-leaning catalog coverage, traffic source mix, geo distribution, bounce rate, domain WHOIS, how-to-use steps.
- Manual run (verification pending)
- FuturepediaSunset
- Legacy entries kept for historical continuity. New ingest is disabled.
- Disabled (30-day sunset)
Activity signal sources
Independent public probes that feed the Mindber Activity Score. Weighted signals sum to 100%. Supplemental signals are collected but do not carry a standalone score weight.
- Public traffic estimates
- Estimated monthly visitors and 30-day change. Sourced via the Toolify and AIbase ingest paths above, not a direct third-party API.
- Daily–weekly by tier
- 30%
- Social — YouTube Data API v3 + LinkedIn
- Days since last public video or post on linked YouTube / LinkedIn channels. YouTube activity is fetched via YouTube Data API v3 (read-only, public data only). X / Twitter was retired 2026-05-01 after both syndication endpoints were disabled.
- Weekly
- 20%
- Public ATS endpoints
- Open-roles count from Greenhouse, Lever, and similar public hiring boards plus careers pages.
- Weekly
- 15%
- HTTP homepage probe
- HTTP reachability, content-hash change detection, page title, TLS health.
- Daily–monthly by tier
- 15%
- Public DNS (A / MX / NS)
- DNS resolution and record presence via public DNS lookups. RFC 1035 hygiene checks.
- Daily–monthly by tier
- 10%
- Public pricing pages
- Vendor pricing-page reachability, number of detected prices, change-over-time deltas.
- Daily–monthly by tier
- 10%
- GitHub API (public)
- Public repository count, commit activity (last 90 days), star trajectory, and open-issue count from linked GitHub organisations. Read-only access to public data only via GitHub REST API v3.
- Weekly
- Supplemental
- Reddit public API
- Subreddit subscriber count and post frequency for tool-specific communities. Public subreddit data only — no user data, no private posts.
- Weekly
- Supplemental
- RSS / Atom feeds
- Blog and changelog publication frequency from tool-linked RSS/Atom feeds. Used to detect content freshness independently of social activity.
- Daily
- Supplemental
Retired sources
Sources that previously appeared on this page but have been removed. We list them explicitly so the historical claim is auditable.
- X / Twitter
- Retired 2026-05-01. Both X syndication endpoints were disabled, making non-API collection unreliable. The social signal now reads only LinkedIn and YouTube.
Ingest policy
- Every probe respects
robots.txtper RFC 9309 before fetching, and pauses the source on a hard reject. - Engagement fetches identify themselves as
Mindber-LivenessBot/1.0with a contact URL — vendors can block us cleanly if they prefer. - Vendors who want any source removed from their product page can file a takedown or claim their profile to control what is shown.