Zero CVEs is not the same as no risk¶
A device with no CVEs in BreachSpider is not shown to be free of exposure, and it never means the device is secure. It means one precise thing: no published CVE that our sources hold matched the product and version you recorded. This page explains the ways an empty result can happen, and how BreachSpider tells them apart.
Why an empty list can be misleading¶
1. The product was not identified¶
If the text in your inventory does not match a catalog product, no CVE check runs at all. An empty result here means unknown, not "nothing found". BreachSpider marks the device unresolved, flags it needs review and lists the closest catalog products. Fix the vendor and model text (use the names on the device label or the vendor's product page) and check again. The Data quality page lists every such device.
2. No version was recorded¶
Without a version, every CVE for the product is listed at product level (PRODUCT_WIDE) and none is confirmed for your device. Add the installed version to get a device-specific answer.
3. No source publishes version data for the product¶
Some products have no CVE data behind them at all, or only records with no version limits. The coverage field says no_cpe_data (undetermined) or partial (CVEs apply, versions cannot be checked). Zero findings here means the sources are silent, not that the device is fine. Check the vendor's own advisories.
4. The vendor stopped supporting the product¶
This is the most important case in industrial plants. When a product reaches end of support, the vendor usually stops investigating and publishing vulnerabilities for it. New weaknesses are still found; they are simply not published as CVEs for that product. An old controller can therefore show fewer CVEs than its supported successor, while being more exposed.
That is why BreachSpider shows lifecycle next to findings. A device past end of support with zero CVEs is a replacement decision (a capital expense), not a patching task (an operating expense). See Lifecycle and firmware currency.
5. Vulnerabilities described only in prose¶
NO_MATCHING_RANGE (the product has range data, but none contains your version) is determined from structured sources: CVE records and machine-readable vendor and government advisories. A vulnerability described only in the text of an advisory would not appear.
6. Not yet published¶
A CVE published today reaches BreachSpider with the next import. The page and every report show when each source was last read.
What a trustworthy "no findings" looks like¶
You can rely on an empty result only when all of these hold:
- The product is
resolvedwith HIGH or MEDIUM confidence. - A version was recorded and compared.
- Coverage is
covered(version-bounded data exists for the product). - The device is supported by its vendor, or you have accounted for it not being supported.
Even then, the statement is limited to published CVEs in our sources. It says nothing about configuration, network exposure, weak passwords or malware.
How BreachSpider words it¶
- In the web app: "A device with no findings is not shown to be free of exposure; it means no listed CVE matched its recorded product and version."
- In the API:
needs_review: truewhenever an empty or partial result must not be relied on, with the warning that explains why. - Over MCP: each device has an
assessmentsentence, and assistants are told never to present an empty, unresolved or partial result as free of risk, and to repeat the assessment in their answer.