Public PoC CVEs: Pull All 47,926 SSVC poc Records

CISA-ADP marks 47,926 CVE records with SSVC Exploitation poc, meaning a public proof of concept or a well-known exploitation method. We pulled them from the cvelistV5 corpus, joined EPSS and KEV, and show what the label does and does not tell you, with a script that writes the full list to CSV.

A brass key hanging on a ring from the keyhole of a white door, above a black iron handle
Photo by Jaye Haych on Unsplash

Short answer: CISA’s Vulnrichment program writes an SSVC Exploitation value into the CISA-ADP container of the CVE records it scores. The value poc means a public proof of concept or a well-known exploitation method exists. In the cvelistV5 snapshot of 2026-11-18 00:00 UTC, 47,926 records have it. That makes it the largest machine-readable list of “public exploit available” CVEs you can download for free, but it is a capability label, not a priority queue: only 2.6% of those CVEs score 0.1 or higher in EPSS.

This article shows how to pull the list, what the counts look like next to KEV and EPSS, and how to cut it down to something a team can act on. For what each SSVC field means and how CISA’s decision tree turns them into Track, Track*, Attend or Act, see our Vulnrichment SSVC guide. For where this fits in a patch decision from CVE record to ticket, see the vulnerability prioritization workflow.

What poc means, in the spec’s words

The SSVC project at Carnegie Mellon SEI defines the Exploitation decision point in exploitation.py. It has two versions, and the meaning of the middle value narrowed between them.

VersionValueDefinition (verbatim)
1.0.0PoC“One of the following cases is true: (1) private evidence of exploitation is attested but not shared; (2) widespread hearsay attests to exploitation; (3) typical public PoC in places such as Metasploit or ExploitDB; or (4) the vulnerability has a well-known method of exploitation.”
1.1.0 (latest)Public PoC“One of the following is true: (1) Typical public PoC exists in sources such as Metasploit or websites like ExploitDB; or (2) the vulnerability has a well-known method of exploitation.”
bothActive“Shared, observable, reliable evidence that the exploit is being used in the wild by real attackers; there is credible public reporting.”
bothNone“There is no evidence of active exploitation and no public proof of concept (PoC) of how to exploit the vulnerability.”

CISA’s Vulnrichment README describes a poc value as “indicating there was a public proof-of-concept available at the time of analysis”. Two practical points follow:

  • “Well-known method of exploitation” counts. A poc label does not guarantee a working exploit sits on GitHub or Exploit-DB. Some records get it because the bug class is trivially exploitable.
  • “At the time of analysis.” The SSVC block carries a timestamp. A none scored a year ago says nothing about the PoC that appeared last week.

The CVE record does not say which Exploitation version CISA applied. Every CISA-ADP SSVC block in our snapshot carries "version": "2.0.3", which is a version string for the SSVC block, not for the decision point.

Where it sits in the record

The value lives in the CISA-ADP container, as one of three single-key objects in options. This is the example from CISA’s own README (CVE-2024-25522):

"options": [
  { "Exploitation": "poc" },
  { "Automatable": "yes" },
  { "Technical Impact": "total" }
]

The path is containers.adp[] where providerMetadata.shortName is CISA-ADP, then metrics[].other where type is ssvc, then content.options. The CVE ADP containers guide covers the container itself.

The numbers: poc next to KEV and EPSS

We counted every CISA-ADP SSVC block in the cvelistV5 release 2026-11-18_all_CVEs_at_midnight.zip.zip, then joined the KEV catalog (version 2026.11.18, 1,786 entries) and the EPSS scores for 2026-11-18 (model v2026.06.15). Every SSVC record had an EPSS score.

ExploitationRecordsShareIn KEVMedian EPSSEPSS ≥ 0.1EPSS ≥ 0.5
none173,13677.7%30.003612,026598
poc47,92621.5%10.004781,231308
active1,7870.8%1,7820.535351,348916
Total222,8491,786

Three things stand out.

1. poc is not a queue for KEV. You might expect a pool of poc CVEs waiting to be added to KEV. The current data cannot show that, because CISA moves a record to active when it joins the catalog. 1,782 of the 1,786 KEV entries read active. The other four (CVE-2024-10441, CVE-2026-40217, CVE-2026-90817 and CVE-2026-90436) all have a KEV dateAdded of 2026-11-18, the same day as the snapshot, so their CISA-ADP blocks had not caught up yet. The single poc record in KEV is one of them. To measure how many poc CVEs later become KEV entries you need two points in time; our CVE exploitation alerts post diffs snapshots that way.

2. active and KEV are not identical. Five records read active but are not in the current KEV catalog: CVE-2023-24932, CVE-2022-21894, CVE-2021-30120, CVE-2021-30119 and CVE-2021-21983. Treat the KEV JSON feed as the authority for “known exploited” and the ADP value as a second opinion.

3. Most poc CVEs score low in EPSS. The poc median (0.00478) is only slightly above the none median (0.00361). EPSS estimates the probability of exploitation activity in the next 30 days, and a published PoC raises that only somewhat on its own. Read our EPSS feed guide before comparing these numbers to a percentile threshold.

There is also a coverage gap to keep in mind. Of the 17,558 CVEs with an EPSS score of 0.1 or higher on 2026-11-18, 12,953 have no CISA-ADP SSVC block at all. CISA scores records as they are published and has not scored the back catalog, so an absent poc does not mean “no PoC”.

Slice poc by the other two decision points

Exploitation is one of three values CISA writes. Splitting the poc set by the other two gives smaller groups with different EPSS profiles:

AutomatableTechnical Impactpoc recordsMedian EPSSEPSS ≥ 0.1
yestotal5,3280.00891394
nototal9,7010.00628377
yespartial9,8870.00516165
nopartial23,0100.00385295

The poc / yes / total slice (11.1% of poc records) has about twice the median EPSS of the full poc set. It is also the combination CISA’s own decision tree treats most seriously short of active exploitation: our Vulnrichment guide shows it at Attend for high mission and well-being impact.

The highest-scoring poc CVEs are not in KEV. CVE-2023-27372 (EPSS 0.99682) and CVE-2025-1974 (0.99400) are both poc / yes / total and absent from catalog version 2026.11.18. In the 2026-11-18 data, 308 poc CVEs score 0.5 or higher. That small set is where poc earns its keep: a public PoC, a high EPSS score and no KEV entry yet.

Pull the list yourself

The script below reads the cvelistV5 daily release zip and an EPSS daily file, and writes every poc CVE to CSV. It uses only the Python standard library.

Inputs:

  • CVE corpus: the <date>_all_CVEs_at_midnight.zip.zip asset from a cve_<date>_0000Z release of CVEProject/cvelistV5 (about 660 MB; it holds an inner cves.zip). Some days the asset is missing; take the previous day’s.
  • EPSS: a daily epss_scores-YYYY-MM-DD.csv.gz from the empiricalsec/epss_scores repository. Its first line is a #model_version comment, which the script skips.
#!/usr/bin/env python3
"""List CVEs that CISA-ADP marks SSVC Exploitation 'poc', joined to EPSS.
usage: public_poc.py <cvelistV5 *_all_CVEs_at_midnight.zip.zip> <epss_scores-YYYY-MM-DD.csv.gz> > poc.csv"""
import csv, gzip, io, json, sys, zipfile

def cisa_ssvc(record):
    for adp in record.get("containers", {}).get("adp", []) or []:
        if adp.get("providerMetadata", {}).get("shortName") != "CISA-ADP":
            continue
        for metric in adp.get("metrics", []) or []:
            other = metric.get("other", {})
            if other.get("type") == "ssvc":
                content = other.get("content", {})
                opts = {}
                for d in content.get("options", []):
                    opts.update(d)  # options is a list of one-key objects
                return content.get("timestamp"), opts
    return None, None

epss = {}
with gzip.open(sys.argv[2], "rt") as f:
    next(f)  # first line is "#model_version:...,score_date:..."
    for row in csv.DictReader(f):
        epss[row["cve"]] = (row["epss"], row["percentile"])

outer = zipfile.ZipFile(sys.argv[1])
inner = zipfile.ZipFile(io.BytesIO(outer.read(outer.namelist()[0])))
out = csv.writer(sys.stdout)
out.writerow(["cve", "datePublished", "ssvc_timestamp", "automatable",
              "technical_impact", "epss", "epss_percentile"])
for name in inner.namelist():
    if "/CVE-" not in name or not name.endswith(".json"):
        continue
    rec = json.loads(inner.read(name))
    ts, opts = cisa_ssvc(rec)
    if not opts or opts.get("Exploitation") != "poc":
        continue
    cve = rec["cveMetadata"]["cveId"]
    e, p = epss.get(cve, ("", ""))
    out.writerow([cve, rec["cveMetadata"].get("datePublished", ""), ts,
                  opts.get("Automatable"), opts.get("Technical Impact"), e, p])

We ran it with Python 3 against the 2026-11-18 release and EPSS file: it took about 37 seconds and wrote 47,926 rows. It loads the inner zip into memory, so give it at least 2.5 GB of RAM (peak resident memory was about 2.2 GB in our run). To keep the list current between full runs, apply cvelistV5’s hourly changes as described in our delta.json and deltaLog.json guide.

Then filter. For example, poc / yes / total with EPSS at or above 0.1 that is not in KEV:

import csv, json
kev = {v["cveID"] for v in json.load(open("known_exploited_vulnerabilities.json"))["vulnerabilities"]}
rows = [r for r in csv.DictReader(open("poc.csv"))
        if r["automatable"] == "yes" and r["technical_impact"] == "total"
        and float(r["epss"] or 0) >= 0.1 and r["cve"] not in kev]
print(len(rows))

On the 2026-11-18 data this prints 394, since no poc / yes / total record is in KEV. The 0.1 cut-off is an example, not a recommendation; pick a threshold you can staff.

How to use poc in a patch decision

  • Never as the only gate. 47,926 CVEs is a backlog, not a priority list. Match the list against your inventory first; the share that applies to you will be far smaller.
  • Rank inside it. Within matched poc CVEs, sort by Automatable and Technical Impact, then EPSS, then your exposure (internet-facing first).
  • KEV beats poc. A KEV entry means confirmed exploitation and, for US federal agencies, a due date. Check the KEV feed before the poc list.
  • Watch for upgrades. A move from none to poc, or from poc to active, is the event worth an alert. Store the value you last saw and compare.
  • Check the timestamp. A none with an old SSVC timestamp on a CVE that is now in the news may simply be stale.

Limits of this article

  • The counts describe one snapshot: cvelistV5 at 2026-11-18 00:00 UTC, KEV version 2026.11.18 and EPSS scores of 2026-11-18. They move daily.
  • A current snapshot cannot tell you how often poc CVEs become KEV entries, because CISA rewrites the value to active. That needs historical snapshots.
  • CISA-ADP covers records CISA has scored, mostly those published or updated since 2024. Older CVEs and many 2022-2023 ones have no SSVC block.
  • We did not check whether the PoCs behind individual poc labels work, or where they are published.

Sources and verification

Important claimSourceStatus
Exploitation definitions, versions 1.0.0 and 1.1.0CERTCC/SSVC exploitation.pyVerified (verbatim)
poc means a public PoC “at the time of analysis”; CVE-2024-25522 examplecisagov/vulnrichment READMEVerified (verbatim)
222,849 SSVC records; 47,926 poc, 1,787 active, 173,136 nonecvelistV5 release, 2026-11-18 midnight, counted with the script aboveVerified (computed)
KEV: 1,786 entries; 1,782 active; four same-day additions; five active not in KEVcisagov/kev-data, catalog 2026.11.18, joined to the corpusVerified (computed)
EPSS medians and threshold countsempiricalsec/epss_scores, 2026-11-18 file, model v2026.06.15Verified (computed)
EPSS estimates exploitation activity over the next 30 daysempiricalsec/epss_scores READMEVerified

Frequently asked questions

What does Exploitation poc mean in a CISA-ADP SSVC block?

It is CISA's assessment at analysis time that a public proof of concept exists or the flaw has a well-known exploitation method. The current SSVC definition (Exploitation 1.1.0) reads: "Typical public PoC exists in sources such as Metasploit or websites like ExploitDB; or (2) the vulnerability has a well-known method of exploitation." It does not mean exploitation in the wild; that is active.

How many CVEs have a public PoC according to CISA?

In the cvelistV5 snapshot of 2026-11-18 00:00 UTC, 47,926 CVE records carry a CISA-ADP SSVC block with Exploitation poc. The number changes daily as CISA scores new records and moves some to active.

Do poc CVEs end up in the KEV catalog?

When a CVE reaches KEV, CISA-ADP moves its Exploitation value to active, so the current poc set is almost entirely non-KEV. In our snapshot only one poc record was in KEV, and it had been added to KEV that day. To find poc CVEs that later became KEV entries you need history, for example a diff of cvelistV5 snapshots.

Should I patch every CVE with a public PoC first?

Not on that signal alone. 47,926 records is too many to treat as urgent, and most score low in EPSS. Combine poc with Automatable, Technical Impact, your exposure and EPSS, and put KEV entries ahead of all of them.

Vulnerability ManagementSsvcCisa VulnrichmentCisa-AdpEpssKevCvelistv5Proof of Concept