CISA KEV Inventory Crosswalk & Changes

Turn a supplied CVE watchlist into source-linked evidence for each asset reference. This Actor is for vulnerability analysts who need structured joins and repeatable comparisons outside their existing vulnerability-management platform.

Run the synthetic sample, then replace its opaque references and CVE IDs with your own approved inventory. The Actor does not connect to, scan or inspect those assets.

What you receive

One dataset row per unique, valid asset/CVE pair: KEV membership, source catalog version and release time, source hash, CISA record fields and the change against your supplied previous snapshot. Different assets associated with the same CVE remain separate rows. Repeated identical pairs within one run produce one row.

Changes are BASELINE, UNCHANGED, ADDED_TO_KEV, NO_LONGER_IN_CURRENT_CATALOG or KEV_RECORD_CHANGED. A missing current catalog entry is not proof that a vulnerability was fixed, that an asset is safe, or that CISA formally withdrew a record. Supplied asset associations and historical snapshots are not independently certified. CISA dates are reported as source fields; their applicability to your organization is not determined.

Read OUTPUT first. EVALUATED means the declared pairs were evaluated; PARTIAL means invalid identifiers or a delivery limit left gaps. UNKNOWN means source coverage could not be established, never a clean result. A catalog more than seven days old, over five minutes in the future, or older than your supplied evidence produces UNKNOWN and no new paid rows. This conservative freshness policy can withhold a catalog even when CISA has simply published no update.

Price

0.005 USD per delivered unique asset/CVE evaluation, with no Actor start fee. Both listed and unlisted pairs are evaluations. Unchanged results are charged on another run. Invalid identifiers, duplicate pairs within a run and failed source evaluation produce no paid row. At most 10,000 pairs are accepted: a full run costs at most 50 USD in Actor event charges. Set your Apify run charge limit before starting; platform/account terms apply. The underlying CISA feed is free. You pay for the inventory join, comparison and delivered evidence.

Inputs and repeated runs

Input is at most 5 MiB. Supply 1–10,000 objects containing only asset_ref and cve_id. An asset reference is 1–80 letters, numbers, underscores or hyphens. CVE IDs are normalized to uppercase. Invalid CVE formats appear in the private UNRESOLVED record and produce no dataset row.

For changes, paste the previous private NEXT_SNAPSHOT array into previousSnapshot. Without it, the run establishes a baseline. After a capped run, NEXT_SNAPSHOT advances only rows actually delivered and preserves supplied prior evidence for undelivered current pairs. Re-running all pairs evaluates and charges them again; submit only the pairs you intend to reevaluate. DELIVERY_UNKNOWN or DELIVERY_RECONCILIATION_REQUIRED requires inspecting the existing dataset and charge record before another run. A resumed run does not automatically replay an ambiguous delivery.

Privacy and limits

The two input arrays use Apify encrypted input fields. Before reading input or writing output, the Actor restricts the run, default dataset and key-value storage, then checks that unauthenticated requests are denied. If access remains available or cannot be checked, it stops. Output links require authorized access; this Actor does not expose signed public download links. Apify and authorized account users still process the data. Use opaque identifiers; do not submit credentials, hostnames, personal information or confidential advisory text. Storage retention follows your Apify account settings; delete the run's data when no longer needed.

Only the fixed official CISA GitHub mirror is fetched, with no redirects and a 5 MiB source limit. No arbitrary URL collection, exploit checks, patch verification, legal interpretation, automatic remediation, alerts or claims of comprehensive vulnerability coverage are included. If your existing platform already supplies this inventory join and history, its native workflow may be sufficient.

Source: CISA KEV official repository, CC0. This product is independent of CISA. For reproducible errors, use the Actor Issues tab with a run ID and synthetic reproduction; do not post customer inventory. No response-time or security-outcome guarantee is made.