Security slang and NVD prose rarely share a word. "sqli" found 843 CVEs
where "sql injection" found 12,719, "lpe" 80 against 1,311 for "privilege
escalation". A term now expands to its phrases and an entry matches on any
of them, with the best-scoring form counted.
A vulnerability searched by the name it is known by leads with the record
it belongs to, whatever the prose and the repository names say: ghostcat
put a different record first, moveit a newer one. Forty names are mapped.
A term with no word in the index at all is nearly always a typo. The
closest word one edit away stands in, weighed by how many records use it,
and the status line says which: citrx finds citrix, fortinte fortinet,
l0g4j log4j.
CVE pages carry BreadcrumbList markup for the home, year and block trail,
and the site publishes an OpenSearch descriptor so a browser can add it
as a search engine from the address bar.
A CVE List V5 record names what it is about: a title such as "PAN-OS:
Arbitrary File Creation Leads to OS Command Injection Vulnerability in
GlobalProtect", a vendor and its products. None of it was kept, so a query
for pan-os or prisma access only found records whose prose happened to say
so. Every record now opens with the title in bold and a Vendor badge per
affected vendor, refreshed from a local CVE List checkout in one pass and
kept current by the daily sync for the records it fetches. Only the lines
above the description move; descriptions and PoC sections are untouched.
The pass also settled the heading link on cve.org and trimmed the stray
space some version badges carried.
The site index carries title, vendor and product for the 40 percent of
indexed CVEs whose records have them. Search scores a hit in those above
any position in the description, results show the title as a headline,
and each page puts it under the CVE id, in the browser title, the meta
description and an Affected row, with the block hubs listing it in place
of the description's opening.
The 35 MB metadata file the homepage needed before anything could be
filtered is split in two for the page: the CVSS rows load with the rest,
the advisories follow the index instead of competing with it for the
connection. cve_metadata.json stays whole for anyone reading the API.
Every keystroke scanned 28 MB of description text, and for terms of four
characters or more the loose regex then ran against every entry that did
not match, about 80,000 regex tests a keystroke: 100 to 230 ms on a
desktop, several times that on a phone. The index now maps each word to
the entries carrying it, built in slices after the data lands so typing
never waits on it, and a term is a prefix lookup plus a scan of the
candidates alone: 7 to 40 ms measured on the same queries. The loose form
of a term, "log-4j" for log4j, is only found by the full pass, which now
runs when the indexed pass comes back nearly empty and not otherwise. The
severity and source of an entry are cached as well, so a filter pass over
the whole index takes 20 ms instead of half a second.
Within a score tier the order fell straight through to the year, which put
five 2026 records above regreSSHion for "openssh" and kept Log4Shell off
the first page of "log4j". Ties now go to the evidence first: CISA's
listing, how many PoCs exist, how likely exploitation is, and only then
the year. A switch on the results panel offers newest and most PoCs
instead, and rides in the URL like the query and the filters. The title,
vendor and product of a record score above its prose once the records
carry them.
The two filter menus hid their state behind ALL and took two clicks per
value. Each severity and source is now its own chip in the row, always
visible: the first click narrows to that value, further clicks widen the
set, and switching the last lit chip off clears. Active chips use the
colours the result rows already use for the same words, and the
selection rides in the URL as sev= and src=, so a narrowed view can be
linked like a query.
Standalone CVE pages carried a smaller search box beside the wordmark
that read as another site's. They now open with the homepage bar itself,
same height, type and placeholder, under the wordmark with the index
count on the right. Enter still submits to /?q= because the pages are
static, and the form has a real submit button for screen readers.
Every writer now has a timeout, so a hung download cannot hold the
repository-updates lock for six hours and cancel the runs queued behind
it. The link audit moves from one 12,000-and-12,000 batch a week, which
measured an hour under that lock, to a seventh of it every day.
The social card PNG is drawn at deploy time and no longer committed: it
changed about four times a day at 68 KB a copy. nuclei.json stops
mirroring EPSS scores; every indexed CVE already has one in epss.json,
the site never fell back to the copy, and the daily recomputation
rewrote 3,400 lines of it each night.
The metadata sync retries GitHub advisory calls on 403, 429 and 5xx
instead of failing on the first one, and refuses to publish a year
shard that an NVD feed would shrink by a tenth or delete outright.
The per-CVE pages read as a different site and disagreed with search about
their own contents. Both traced to the page generator having grown its own
copy of rules the browser already had.
Security:
- JSON-LD was serialised with json.dumps alone, which leaves < > and & raw. A
description containing "</script>" ended the element and turned the rest into
live markup. 78 descriptions in the corpus already carry that sequence; only
the 155-character meta trim kept them out of the payload. Escape to \u00XX.
Correctness:
- The PoC count summed only the repository column, so 19,940 pages announced
"0 public PoC exploits" directly above the ExploitDB and Metasploit entries
they were listing, in the title and meta description too. Count every source
the page renders, matching POC_FIELDS in logic.js.
- The severity chip took the highest score while search took the first, so
18,434 CVEs showed one severity in the results and another on the page. Both
now take the primary assessment, with the same Nuclei fallback search uses.
- Identical CVSS rows printed two and three times on 13,340 pages. Dedupe on
(version, score, vector) and label each survivor with its assessor from
row[4]; row[5] is the assessment type and left 129 pages showing two rows
both marked "Secondary".
- repo_from_url accepted any host containing "github" and kept .git suffixes,
so a clone URL and its plain form counted as two sources on the page and one
in search. 273 CVEs disagreed. Match repoFromUrl in logic.js exactly.
- 41 CVEs nobody has scored were unreachable by search, including by exact ID:
the severity filter tested membership of a list naming only the four scored
bands, so ALL excluded them. ALL now means all.
- EPSS percentiles printed a hard-coded "th": 51th, 62th, 93th on 21,100 pages.
- Search reported "1 PoCs", and the empty-results branch returned before
updating the live region, leaving it reading "loading index..." forever.
Deploy safety:
- A failed EPSS fetch returned {} and published it, silently erasing every
score from 82,000 pages. It now raises, as does a feed covering under 90%
of the index.
Presentation:
- CVE pages never loaded the webfonts the homepage loads, so all 82,468 fell
back to system fonts. This is why they looked unlike the rest of the site.
- 404.html referenced four CSS classes that no longer existed and rendered
unstyled, and its meta refresh threw a reader out of a search after two
seconds. It is now generated from the same constants and header as every
other page, and offers the search box instead.
- The homepage and the generated footer credited different sources. Both now
come from brand.SOURCES_LINE, and the build refuses to publish a homepage
that has drifted from it.
- Three aria-labels sat on div elements with no role, where they are ignored.
SEO:
- 61,936 pages filled "Related" with numerically adjacent CVEs, which are not
related to anything. Keep only shared-author relations; 20,532 pages have one.
- Sitemap lastmod described the CVE record, so a CVE that gained forty PoCs
still advertised a date from 2021. It now reflects the page, and each shard
carries its newest page's date rather than today's.
- TechArticle is not a type Google draws a rich result from, and its dates
claimed the CVE record's publication as the page's. Use WebPage with the
vulnerability as mainEntity.
- Curated sources and the CVE and NVD records no longer carry nofollow; the
crowd-sourced repository list still does.
82,000 CVEs sat behind one URL. Each of them is a query somebody types, and the
pages answering those queries belonged to other people: searching
"CVE-2026-72898 poc" returns SentinelOne, Horizon3, Wiz and BishopFox, all of
them serving one page per CVE.
Pages are flat files at /CVE-2026-72898. No /cve/ prefix, no extension, no
trailing slash. GitHub Pages resolves an extensionless request to the .html
beside it, checked against the live site rather than assumed.
Each page carries what the index already knows and nothing invented: every CVSS
assessment with its vector, EPSS with percentile, KEV date and ransomware flag,
Nuclei severity and CWE, and the PoC links grouped by source with stars and last
push. 94% of them carry three or more of those signals.
Discovery works two ways. sitemap.xml is now an index over one shard per year,
which fits the 50,000 URL cap on the split the data already has and lets a
crawler refetch only the year that moved. Search results link to the CVE page
instead of straight out to NVD, which moves to a chip beside CVE.ORG, and each
page links six related CVEs picked by shared PoC author.
Rendering all of them takes 20s and adds nothing to the repository: the payloads
they read are already gitignored and built in CI, so the pages go into the Pages
artifact and git never sees them.
The site called itself three different things: CVE PoC Hub in the title, CVE
Radar in the h1 and the social metadata, and cve on GitHub. Google had no
consistent site name to settle on, so none of the mentions compounded. CVE
Radar was taken anyway, by SOCRadar's free tool and by cveradar.com.
Every user-facing name and URL now comes from scripts/brand.py, so the next
move is one edit rather than a sweep through the tree.
Also repairs the parts of the site a crawler could not use:
- robots.txt and sitemap.xml were both 404. scripts/build_seo.py writes them,
wired into the Pages job. trending_poc.json, kev.json and stats.json stay
crawlable because the first paint needs them; the five bulk payloads are
disallowed, which is 13.2 MB per crawl spent on data no crawler queries.
- Search state lives in ?q= now. Results were unlinkable, so nobody could cite
one and there was no route for a SearchAction to declare. replaceState
rather than pushState, so typing does not bury the back button.
- A WebSite JSON-LD block carries the name Google prints as the site name.
- The footer carries the MITRE trademark notice.
Four gaps, measured against the index before any of this was written.
ExploitDB and Metasploit are folded in the way nuclei templates were. ExploitDB
covers 25,066 CVEs and adds 30,327 links across 17,237 entries; Metasploit
covers 3,178 and adds 3,520 across 2,550. Together with the templates the index
goes from 101,809 CVEs carrying a PoC to 103,945. Neither archive skews recent,
which is why nuclei went first, but both are the difference between an old CVE
having nothing and having something that runs. Each keeps its own section and
its own field, and each leads the list ahead of the repositories: a module, an
entry and a template all run as they stand, without reading somebody's code.
EPSS now covers the index rather than the 4,274 CVEs a template happened to
rate. FIRST's daily feed scores 103,783 of 103,945, so a result can say how
likely a flaw is to be exploited in the next thirty days, not only that code
exists for it. It is fetched during the site build and never tracked here: a
daily rescore of 366,000 rows would be nothing but churn.
The trending tables also stop hiding what they exist to show. The star floor
excluded 1,059 of the 1,312 2026 PoC repositories pushed this quarter, and a
repository published this morning has no stars, so the newest row on the front
page was six days old. A separate lane drops the floor entirely and pays for it
with a ten day window, a cap of ten rows, and the same paperwork check the rest
of the table uses. The newest row is now hours old.
A nuclei template is a runnable check for one CVE, which makes it a proof of
concept whether or not anyone published a repository for it. Measured against
the index: 3,952 CVEs gain a template link and 858 of those had no PoC at all,
584 of the additions falling in 2024 or later. That last figure is the point.
ExploitDB carries more CVEs overall and Metasploit is better weaponised, but
both skew historical; nuclei is where coverage of a recent CVE actually shows
up first.
Templates also carry the ratings the entries never had. Severity, CVSS with
its vector, EPSS with its percentile and CWE now sit in nuclei.json for the
site to read: 4,193 of 4,418 templates state a CVSS score and 4,274 an EPSS
score. A result can finally say how bad a flaw is and how likely it is to be
used, rather than only that somebody wrote code for it.
Templates are kept in their own section and their own field. They are not
somebody's repository and should not be ranked, starred or credited as one,
but they do lead the PoC list, being the one link that runs as it stands.
The index files are also revalidated rather than reused from cache. They are
rebuilt hourly and a returning visitor was holding whatever it downloaded
first, which cost a stale index here until the copy was thrown away by hand.
Landing on the page and typing did nothing, so the first thing anyone tries
was lost and the field had to be clicked first. It is focused on load on
viewports wide enough to imply a keyboard, and typing anywhere else moves
focus there.
Focus moves during keydown rather than on the character itself, so the
keystroke that triggered it still lands in the field: typing gitea from
anywhere on the page searches for gitea, not itea. A lone slash focuses
without typing itself, and escape clears the field and steps back out.
The page and the README had drifted into looking like two projects. The site
now opens on the same card the README banner draws, built as markup rather
than a second copy of the image, so it fills itself from the index totals the
page already loads and cannot go stale.
Contrast was the real problem. --neutral sat at 4.12:1 against the canvas,
under the 4.5 floor, and it carried the search placeholder, the star glyph and
every separator; descriptions were rendered in the muted token at 6.15:1 while
doing the work of body copy. Descriptions move to a subtle token at 12.26:1,
muted and neutral both come up, and links go to the brighter accent at 7.49:1.
The column headers were 9.5px mono under a tenth of an em of tracking, which
no contrast ratio was going to rescue.
Known-exploited CVEs are marked in the trending table too, not only inside a
result, since which row to read first is the one thing the table could not
say. The README swaps the fire emoji for the same KEV pill, served from the
repository so every flagged row points at one cached file.
The search field no longer repeats the totals now that the hero carries them,
and keeps its slot for the match timing.
Every colour is now a Primer dark_default token. The header band, its four
capability pills and the heading above the trending table are gone, so the
search field is the first thing on the page and the switcher labels its own
table. Rows tighten to 10px, descriptions clamp to two lines, and the PoC
list loses its box for a hairline rule.
Blue is the only thing you can click. Amber means 500 stars or more and red
means known-exploited, and nothing else carries a hue: the star glyph is
dimmed at every value so the column reads as numbers rather than symbols.
The handoff's placeholder wanted an em dash. It reads "Search CVE, vendor,
product or keyword, exclude with -windows" with the middot the count line
already uses, and the six em dashes I had left in code comments are gone
too.
An advisory copied from a GitHub Security Advisory opens with "### Impact"
or "### Summary". The section parser treated any heading as a boundary, so
those descriptions ended themselves and reached the site empty. Only the
four headings an entry actually uses end a section now, and leading heading
markers are stripped when the text is rendered.
A trailing slash made exploit-db.com/exploits/41034 and .../41034/ two
different links, which put 53 duplicate rows in front of readers.
github.com also serves attachments and advisories, and 103 of those were
being parsed as owner/repo -- rendering "user-attachments / files" with an
empty star column instead of the PoC archive they point to.
CISA publishes a catalogue of vulnerabilities observed being exploited in
the wild, and 1,111 of its 1,685 entries are CVEs this index already holds
PoCs for. That is the one fact a reader most wants before opening an
exploit, and it was nowhere on the page.
The weekly audit stores the catalogue, the build trims it to the CVEs the
index carries, and a result row shows a KEV chip -- with the ransomware
flag where CISA records one, which covers 352 of them. 37 KB served.
A failed fetch keeps the stored catalogue rather than dropping the badge
off every entry. The chip reuses the header pill's treatment: the palette
still spends green on things you can click and amber on star counts only.
A pass over 104,043 entries costs 25-125ms, which is fast enough to feel
instant once but not nine times in a row while a word is typed. Coalesce a
burst at 160ms, skip single-character queries that match most of the index
without narrowing anything, and total the PoC count during the scan rather
than walking the matches again to display it.
Sorting a CVE's links on stars alone handed the top of Log4Shell to
ThreatMapper, Viper, afrog and Awesome-OSINT-For-Everything: large tools
that mention the CVE somewhere. Rank repositories named after the CVE
first and only order those by stars, which puts CVE-2025-32463_chwoot
above a 2.5k-star general sudo toolkit.
A term no advisory text uses leaves every match tied at the weakest tier,
where newest-first is meaningless; searching log4shell led with text4shell
and then a CVE holding a single incidental link. Break those ties on how
many of the linked PoCs carry the term instead, which is only consulted
when nothing matched a description, so product searches are unchanged.
Also drop two controls that could not do anything: an expander on a
description already short enough to fit, and a footer offering to expand
a list of one.
Space Grotesk for prose, IBM Plex Mono for every identifier, and the flat
hairline-bordered surfaces from the handoff. Green marks what you can act
on and amber only ever means a star count; nothing else is coloured.
Three changes to how the same data reads. Descriptions clamp to three lines
behind an expander, so a kernel CVE no longer pushes the links people came
for below the fold. Raw URLs become owner / repo rows carrying stars and
last-commit age, ranked best first, five at a time. And an empty search box
now keeps the trending PoCs on screen with a TRENDING / RECENT toggle
rather than showing an empty page.
The search grammar, ranking and negation are unchanged.
Typing before the payload arrives searched an empty dataset and reported
no results, then the loader hid the table and left a blank page, so the
query had to be typed a second time. Show that data is loading and run
the pending query once it lands.
Unquoted words were scored independently, so "active directory" matched a
description holding "active session" and "working directory" as strongly
as a real AD advisory, and the newest-first tie-break then put the wrong
one on top. Reward the words appearing together.
Plain substring matching also meant "rce" hit "open source" and "sudo"
hit "pseudo"; require a term to start a word, while still allowing it to
match into one so "word" finds "wordpress". Weight matches near the start
of a description, in coarse buckets so a few characters never outweigh
recency: "log4j rce" now leads with Log4j2 rather than GeoTools.
Scores were boosted by how many PoC links matched, which promoted old
CVEs with large link piles above recent findings; searching WordPress led
with 2015-2018. Rank on match quality alone and break ties by CVE year
and number, newest first.
Only the id and description are needed for every comparison, so PoC text
is derived on demand: preparing 108934 entries drops from 2.9s to 0.6s.
Dropping no-store lets the browser revalidate the 53 MB payload instead
of refetching it on every visit; the 5 KB trending file stays uncached.
Build result rows directly, since String.replace reads $& in the
replacement as a pattern, and escape link text like every other cell.
0xMarcio