How this works

A directory is only useful if it's current. Ours re-checks itself on a schedule, gates real changes behind human review, and shows the receipts.

The pipeline

From a scheduled check to a published, verifiable update.

Step 1

Scheduled checks

Recurring tasks fire on a cadence — re-reading each tool and model page so profiles never quietly go stale.

Step 2

Agents verify

Agents read the source, diff it against the last known state, and record a verification check with a hash and an evidence link.

Step 3

Human review gate

Material changes — a verdict flip, a pricing move — are drafted, then held for a human to approve before anything goes live.

Step 4

Publish with a trail

Approved updates publish with the verification log attached, so every "last verified" badge links back to the check behind it.

A recent run

Real checks from a live profile. Each row re-reads the source and records whether anything changed — open Details for the evidence and content fingerprint behind it.

  • Pricing— No changes

    Automated agent

  • Pricing— Updated

    Automated agent

    Pricing restructured: old Community/Cloud Pro($10/mo 1K tasks)/Cloud Enterprise → new Standard(Free+$5/active flow)/Ultimate(Custom). Integration count now 754+ (was unspecified). New AI agent positioning.

  • Profile— No changes

    Imported at launch

  • Pricing— No changes

    Imported at launch

See the full log on a live profile