CrawlPact

Policy Observatory

Registry Observatory

Every figure below is computed directly from CrawlPact's immutable registry release 2026.09.1, resolved from the same frozen release snapshot production evaluation and historical scan rendering use — never from a live, mutable table, and never from the static crawler directory pages.

Data collected
2026-10-22T10:21:26.120Z
Registry release
2026.09.1
Checksum
9a4f14303d99f24d8ba513d2d158263dd48df3710427fa3ac00065607bb9ed2b
Methodology
policy-observatory-methodology-v1

Crawler and operator counts

Crawler and operator summary counts
Tracked crawler records24
Evaluation-eligible (active, deprecated, replaced)24
Distinct operators9

Purpose distribution

Denominator: 24 tracked crawler records.

Crawler records by declared purpose
PurposeRecords
advertising validation2
agent2
mixed2
research1
search7
training5
unknown1
user triggered4

Lifecycle distribution

Crawler records by lifecycle status
Lifecycle statusRecords
active24

Operators by purpose

Which purposes each operator publishes crawlers for
OperatorPurposes
Amazonmixed, search, user triggered
Anthropicsearch, training, user triggered
Applemixed, training
Common Crawl Foundationresearch
Googleagent, search, training, unknown
Metaadvertising validation, agent, search, training
Microsoftsearch
OpenAIadvertising validation, search, training, user triggered
Perplexity AIsearch, user triggered

This lists which purposes each operator currently documents separate crawlers for — it is not a transparency or trust ranking (no operator quality score exists).

Source-verification freshness

Denominator: 24 tracked crawler records.

Source-verification freshness summary
Records with a recorded verification date24
Never verified0
Median verification age (days)92
Due for re-verification (evaluation-eligible, >180 days or never)0 of 24

Release history

2 published release(s). Each row compares that release to the one immediately before it — evidence-only and editorial-only changes never triggered customer re-evaluation (see methodology).

Registry release history and per-release changes
ReleasePublishedAddedRemovedPurpose changesToken changesEvidence-only
2026.09.1(active)20 Sept 202610000
2026.07.328 Jul 2026230000

Limitations

  • These figures describe CrawlPact's own crawler-identity registry, not a sample of the web — they do not measure how many websites allow or block any crawler.
  • Purpose classification reflects each operator's published documentation as independently reverified by CrawlPact; it does not measure actual crawler behaviour.
  • An operator may run undocumented or unverified crawlers not yet reflected in this registry.

Analytics preferences

CrawlPact uses optional Google Analytics and Microsoft Clarity on public pages to learn which content is useful. Clarity records clicks and scrolling (session replay), with anything you type masked. Neither runs in the app or admin areas, and CrawlPact works the same whether you accept or decline. See the privacy policy.