Skip to main content

Certificate Transparency monitoring

A CT log discovery source watches public Certificate Transparency logs for certificates issued for domains you care about — including certificates issued by other teams, other tooling, or an attacker. Every publicly-trusted certificate must be CT-logged, so this catches issuance that never touched CertAutoPilot.

Availability

CT monitoring relies on the external crt.sh aggregator, which is frequently rate-limited or down. Because of that the CT source type is hidden from the source-type picker by default. For enumerating certificates behind internal, SNI-routed infrastructure, use an AXFR source instead — it is authoritative and needs no external service. CT monitoring remains the tool for catching publicly-issued shadow-IT certificates on domains you own.

How it works​

The CT executor queries crt.sh (the only provider in the current release) once per monitored domain:

  • With Include subdomains on, the query covers the apex and every subdomain (%.example.com wildcard); matching is then re-verified label-safely, so evil-example.com never matches example.com.
  • Fetches are incremental: the source stores the newest CT log entry timestamp (last_ct_entry_time) and later runs skip entries logged before it.
  • Each matching certificate is upserted as a CT observation, deduplicated on (project, provider, provider cert ID, monitored domain) — re-seeing a certificate updates last seen instead of creating a duplicate.

Runs are enqueued as discovery_execute jobs — either by the leader-elected scheduler when the source is due per its schedule interval, or manually via Scan now. If some domains fail (e.g. crt.sh rate-limits), the run completes as partial rather than failing outright.

Configure a CT source​

  1. Go to Discovery → Sources → Add source → type CT Log.
  2. Fill in:
    • Domains — one per line, maximum 50 per source.
    • Include subdomains — default on; off restricts matching to the exact names listed.
    • Provider — fixed to crt.sh.
    • Allowed issuers — optional list of issuer names you expect (e.g. Let's Encrypt, your internal CA). Certificates from any other issuer raise a high-severity finding.
    • Schedule interval — how often the source is polled, in hours; minimum 24.
  3. Save, then trigger the first run with Scan now if you don't want to wait for the schedule.

Observations​

Each observation records the certificate as seen in the log: common name, SANs, issuer, serial number, validity window (not_before/not_after), the CT log entry time, which monitored domain and identifier matched (and whether via CN or SAN), plus first/last seen timestamps.

They surface on the source detail page (Discovery → Sources → source → CT Certificates (N) tab), newest log entries first, up to the 500 most recent.

Findings​

Two finding types feed the Findings page (scope cert, status open):

FindingSeverityRaised when
ct_new_certinfoA certificate not previously observed appears in CT for a monitored domain.
ct_unexpected_cahighAllowed issuers is configured and the certificate's issuer is not in the list.

CT findings are never auto-resolved (certificates never leave a CT log) — they stay open until you Resolve, Acknowledge or Suppress them on the Findings page. A manual resolve is effectively permanent for CT findings: the same CT entry is never re-detected.

note

CT alerting is surfaced through discovery findings, not through the notification event pipeline — subscribe your review workflow to the Findings page, not to a notification rule.

Limits & behaviour​

LimitValue
Domains per source50
Results per domain per fetch500 (truncation is logged — split very busy apexes across sources or rely on incremental catch-up)
crt.sh response cap / timeout10 MB / 30 s
Pause between domain queries500 ms (be gentle to crt.sh)
Minimum schedule interval24 h
crt.sh HTTP 429Domain marked failed for this run; retried on the next run
Detection lag

crt.sh itself lags CT log incorporation, and sources poll on an interval of hours — CT monitoring is an audit control, not a real-time alarm. Trigger the source manually from its detail page when you need an immediate sweep.

API​

CT sources use the standard discovery-source endpoints; observations have a dedicated listing:

POST /api/v1/projects/{projectId}/discovery-sources
POST /api/v1/projects/{projectId}/discovery-sources/{id}/scan
GET /api/v1/projects/{projectId}/discovery-sources/{id}/ct-observations

See also​