Status: GitHub published GHSA-cxc9-87g6-qrj9 on November 21, 2026, rating it critical and classing it as malware in version 2.4.1 of the npm package @wppconnect-team/wppconnect, a library that automates WhatsApp Web from Node.js. GitHub's advisory says any computer that has the package installed or running "should be considered fully compromised" and that its secrets and keys should be rotated from a different computer. The technical detail comes from OSV record MAL-2026-20044 (written by the OpenSSF malicious-packages project, with findings credited to Amazon Inspector and Bitbison) and from a report filed in the project's issue tracker. OSV and the report say version 2.4.1 contains a module that, when the library is imported, collects environment variables, configuration files and WhatsApp session-token files and uploads them to a Telegram bot. The maintainers removed the source on the evening of November 15 and later confirmed that the 2.4.1 artifact contains the reported startup markers. GitHub lists no patched version. As of 01:05 to 01:10 EST on November 22, 2026, npm's registry no longer lists 2.4.1 among the package's versions, and its latest tag points to 2.4.6. No source says how the code reached the repository, who runs the Telegram bot, or whether anyone's data was received. The advisory has no CVE ID.
Executive Summary
This is a supply-chain incident in a package that developers add to chatbots, customer-service tools and WhatsApp automation. The reported behaviour is credential and session theft at the moment the library loads, which makes every machine that ran version 2.4.1 a candidate for secret rotation, whatever it was used for.
What matters for planning:
- Only version 2.4.1 of the library is named by GitHub. The GitHub entry lists the affected range as "= 2.4.1" and no patched version. OSV also lists a second package, @wppconnect/server 2.10.82, as malicious (record MAL-2026-20043), and the maintainers say its artifact contains the reported startup markers. We found no GitHub advisory for the server package when we queried GitHub's advisory API, so treat both as in scope (Rescana guidance).
- GitHub's instruction is full rotation, from a clean machine. GitHub says all secrets and keys on an affected computer "should be rotated immediately from a different computer," and warns that removing the package does not guarantee that all malicious software is gone.
- The behaviour, as OSV and the project issue report describe it, is read-and-upload. It copies environment variables, reads named configuration files from the working directory, takes up to five WhatsApp session-token files, looks up the public IP address and sends one report to a Telegram bot. OSV says it runs once when the library is imported, and at most once every 24 hours per machine. None of the sources we opened describes remote control of the machine. GitHub's wording is broader than the technical descriptions, and we treat the broader wording as the safer instruction.
- The code runs on import, not on install. Because OSV places the trigger in the library's import path, turning off install scripts would not have stopped it (Rescana's reading of OSV's description).
- Nobody has published who is behind it or what was received. No source names an actor, reports victims or says how the code got into the repository, and we name no one. The maintainers say the questions of publishing-account access, revoking the Telegram destination and notifying users were still open when they last replied.
- Scale is unknown. npm's download API reports 9,737 downloads of the package, all versions combined, for the seven days from November 15 to November 21. That is not a count of 2.4.1 installs.
If you run Node.js applications that use WhatsApp automation, find out today whether any lockfile, image or build ever resolved 2.4.1 or server 2.10.82, treat every machine that ran them as exposed, rotate what those machines could read, and log the WhatsApp sessions out (Rescana guidance built on GitHub's and the maintainers' published steps). This advisory is a new post, not a follow-up: no live Rescana post covers this package.
Technical Information
This section reports what GitHub, OSV and the maintainers published. It gives no payload code and no steps for reproducing or hiding the behaviour. No source we opened cites a public proof of concept, none applies because the package itself is the malicious artifact, and this post links none.
GitHub's entry. GHSA-cxc9-87g6-qrj9 is titled "Malware in @wppconnect-team/wppconnect," is typed as malware, rated critical and marked as reviewed by GitHub. It lists CWE-506 (Embedded Malicious Code), has no CVE ID and no CVSS vector, lists the vulnerable range as "= 2.4.1" and lists no patched version. Its description is short and general: it does not say what this package does. The behaviour below comes from OSV and the maintainers.
OSV's description. OSV record MAL-2026-20044 says version 2.4.1 ships a file, dist/utils/diagnostics.js, that is loaded by dist/controllers/browser.js so that it runs when the library is imported. According to OSV and the Amazon Inspector finding it carries, the module:
- records the host name, user name, operating system and working directory;
- copies every environment variable except a short list of shell and npm variables;
- reads
.env,.env.local,.env.production,config.js,config.jsonandormconfig.jsonfrom the working directory, plus the project's dependency list and a directory listing; - takes up to five WhatsApp session-token files from a
tokens/folder; - looks up the machine's public IP address through a public lookup service,
api.ipify.org; - uploads the result as a document to the Telegram Bot API, naming a Telegram bot and a chat that appear in the Indicators section below;
- sends at once on import, at most once every 24 hours per machine, using a marker file in the temporary directory, and ignores errors so that an application carries on normally.
OSV adds that the Amazon Inspector finding describes the Telegram address as built at runtime from pieces rather than written as one string, and that the module is presented as a "runDiagnostics" support helper with Portuguese comments. Rescana's reading: a plain-text search of node_modules for a Telegram address may therefore miss it, so search by file name and package version first.
The maintainers' account. The package's issue tracker holds a report, issue #2968, opened at 23:00 EST on November 15 by a reporter. It says the same code reached a second package, @wppconnect/server 2.10.82, which loads a similar module when the server starts, waits a random 30 to 240 seconds before sending and skips machines that look like CI or sandbox hosts. The report says the library's commits were authored under a maintainer account and describes them as "environment diagnostics" in their commit titles. GitHub's commit pages show the two library commits as unsigned, made at 17:34 EST on November 15, and the release commit for 2.4.1 as made by the project's release bot one minute later. A maintainer replied at 09:44 EST on November 16 that they had downloaded the npm artifacts without running them, verified their registry integrity, and confirmed that "2.4.1 is deprecated and its artifact contains the reported startup markers" while the 2.4.0 artifact does not. The maintainer also said the source removal in pull request #2967 is merged, that these bounded checks do not "establish complete remediation of publishing accounts or affected installations," and that the issue stays open until publishing access, revocation of the reported destination credentials and notification of affected users are confirmed. The reply does not say how the module entered the repository.
What is not known. No source says who operates the Telegram bot, whether Telegram has disabled it, whether any report was received, how the code entered the repository, which npm account or token published 2.4.1, or when and by whom 2.4.1 was removed from the registry. OSV's record still says 2.4.1 is "still npm's latest version"; the maintainers' later comments, the registry metadata and the issue thread show that is no longer true.
Affected Product Versions
GitHub names one version of one package. OSV and the maintainers add a second package. The registry state below is from our own check at 01:05 to 01:10 EST on November 22, 2026.
| Package and version | Status | Source |
|---|---|---|
| @wppconnect-team/wppconnect 2.4.1 | Malicious; vulnerable range "= 2.4.1"; no patched version listed | GitHub Advisory Database GHSA-cxc9-87g6-qrj9; OSV MAL-2026-20044 |
| @wppconnect/server 2.10.82 | Malicious; also reaches machines through its dependency on the library, per OSV; not covered by a GitHub advisory we could find | OSV MAL-2026-20043; project issue #2968; npm deprecation notice on the version |
| @wppconnect-team/wppconnect 2.4.0 | Maintainer says the artifact does not contain the reported markers | Maintainers' reply in issue #2968 |
| @wppconnect/server 2.10.81 | Maintainer says the artifact does not contain the reported markers; npm's deprecation notice on 2.10.82 says to pin 2.10.81 pending a reviewed release | Maintainers' reply in server issue #2734; npm registry |
| @wppconnect-team/wppconnect 2.4.6 | Current latest tag. No source calls it fixed, and GitHub lists no patched version. Our text search of its files found none of the names or strings described above (see below) |
npm registry; Rescana check |
| @wppconnect-team/wppconnect 2.4.2 to 2.4.5 | No entries in the registry's version list or publish-time list. The maintainer says, for 2.4.2 and 2.4.3, that this does not establish whether they were ever published; the repository has tags for 2.4.4 and 2.4.5 | npm registry; maintainers' reply in issue #2968; repository tags |
Four practical points, all Rescana's reading unless marked. First, the registry's publish-time list still records 2.4.1 (published 18:31 EST on November 15), but the version itself is gone: requesting it returns "version not found," and its tarball URL returns not found. Second, removal from npm does not remove copies: private registries, caching proxies, container images, CI caches and old lockfiles can still hold 2.4.1. Third, @wppconnect/server declares the library with the range ^2.3.6 in the versions we checked (2.10.78 to 2.10.82), so what a server install picked up depends on when it ran and on any lockfile, which is why you should check resolved versions and not only package.json. Fourth, for 2.4.6 we downloaded the tarball (its SHA-1 matches the registry's), unpacked it without running anything and searched its files for the module's file name, the marker file name and the words "telegram" and "ipify." We found no match. That is a text search, not a code review, and it says nothing about future releases.
Timeline (EST)
- November 15, 2026, 17:34 EST. Two unsigned commits to the library's repository add a diagnostics module and load it when the browser module starts. These are GitHub's commit timestamps, one second apart; GitHub does not publish when a commit was pushed.
- November 15, 2026, 17:35 EST. The release commit for version 2.4.1 is made by the project's release bot (commit timestamp).
- November 15, 2026, 18:31 EST. npm records the publication of 2.4.1. OSV says it went out through the project's release workflow.
- November 15, 2026, 21:43 EST. A commit to the server project's repository loads a similar module when the server starts (commit timestamp).
- November 15, 2026, 22:09 EST. @wppconnect/server 2.10.82 is published.
- November 15, 2026, 22:44 to 22:52 EST. Removal pull requests are opened and merged in both repositories (server #2733 merged 22:50 EST; library #2967 opened 22:50 EST and merged 22:52 EST). The library's pull request says it does not publish a new version or change versions already in the registry.
- November 15, 2026, 23:00 EST. A reporter opens issues on both projects. The library issue says 2.4.1 is "still latest" on npm.
- November 15, 2026, 23:10 EST. OSV publishes MAL-2026-20044 for the library.
- November 16, 2026, 03:23 EST. A commenter on issue #2968 reports that the
latesttag is back at 2.4.0 and that 2.4.1 shows a deprecation warning. - November 16, 2026, 09:44 EST. A maintainer replies on the issue, confirming the artifact contents and keeping the issue open (details above).
- November 18, 2026, 08:37 EST. npm records version 2.4.6, now the
latesttag. - November 21, 2026, 08:46 EST. GitHub publishes GHSA-cxc9-87g6-qrj9. npm's registry metadata for the package shows the same modification time to within a minute. No source says when 2.4.1 was removed.
- November 21, 2026, 09:30 EST. OSV last modifies the library's record.
- November 22, 2026, 01:05 to 01:10 EST. We re-checked GitHub's entry, the npm registry for both packages, both issue threads, OSV and the Rescana feed. The registry no longer listed 2.4.1 among the versions. The issue thread was still open with the two replies above.
The raw timestamps in GitHub's, npm's and OSV's APIs are UTC, so events between 00:00 and 05:00 UTC on November 16 fall on November 15 in EST. All dates and times in this post are EST.
How to Find Your Exposure
The checks below combine what OSV and the maintainers describe with Rescana guidance, which we mark. None of the sources publishes detection signatures, and GitHub's entry adds nothing beyond its short description.
- Search for the two versions everywhere a package can hide. Look for
@wppconnect-team/wppconnectresolved to 2.4.1 and@wppconnect/serverresolved to 2.10.82 inpackage-lock.json,yarn.lock,pnpm-lock.yaml, software bills of materials, container images, build caches, artifact stores and deployednode_modulesfolders. Include the git history of lockfiles, since an image built on November 15 or 16 may still carry a version that your current lockfile no longer names (Rescana guidance). - Check private registries and proxies. If you run a mirror or caching proxy for npm, check whether it fetched 2.4.1 or server 2.10.82 and whether it still serves them. npm's removal does not reach your copy (Rescana guidance).
- Work out which machines actually ran it. OSV places the trigger in the library's import path and says the server version runs at start. List every host, laptop, container and CI job that imported the library or started the server while a bad version was installed. GitHub's wording covers any computer where the package is installed or running, so start from installs and narrow only with evidence (Rescana guidance).
- Pick the window. The version was on npm from 18:31 EST on November 15. The reporter said it was still the
latesttag at 23:00 EST, and a commenter said it had been moved back by 03:23 EST on November 16. Nobody has published when it was removed. Rescana's reading: yarn and pnpm pick the highest version that fits a range rather than thelatesttag, so for unlocked installs through @wppconnect/server the window may run at least until 2.4.6 appeared at 08:37 EST on November 18. For any machine that installs without a lockfile, start at 18:31 EST on November 15 and widen the window until you can show what your own registry served (Rescana guidance). - Look for the marker files. OSV lists a once-a-day marker file in the system temporary directory,
.wpp-support-diagfor the library, and the project issue #2968 names.wpp-diagfor the server. Their presence suggests the module ran on that machine. Their absence does not clear it, because temporary folders are cleaned and the sources do not say the marker is always written (Rescana's reading). - Look at network logs. Search DNS, proxy, firewall and endpoint logs for a host that contacted
api.ipify.organd thenapi.telegram.orgwithin moments, and for any request whose path contains the bot ID listed in the Indicators section if you inspect TLS. Both services are legitimate and common, so the signal is a server or CI runner that has no reason to talk to Telegram. The library variant sends at once on import and, per the sources, has no stated CI exclusion, while the server variant waits 30 to 240 seconds and skips hosts that look like CI. Retention may not reach back to November 15 (Rescana guidance). - List what was readable. For each machine that ran it, list the environment variables of the process, the
.envfamily and config files in its working directory, the database settings inormconfig.jsonand the contents of anytokens/orwppconnect_tokens/folder. That list is your rotation list (Rescana guidance, from OSV's list of what is collected).
Workaround and Mitigation
GitHub's entry says: "Any computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be rotated immediately from a different computer. The package should be removed, but as full control of the computer may have been given to an outside entity, there is no guarantee that removing the package will remove all malicious software resulting from installing it."
The maintainers and OSV add that anyone who installed or ran the affected versions should treat environment variables, the named files and their WhatsApp sessions as exposed, rotate the credentials the process could reach, and invalidate or log out the sessions. npm's deprecation notice on server 2.10.82 says: "SECURITY INCIDENT: This version contains code that may transmit host environment, configuration and session data to an external endpoint. Do not use. Pin 2.10.81 pending a reviewed release." The maintainers' replies point to 2.4.0 (library) and 2.10.81 (server) as the artifacts without the reported markers. There is no GitHub-listed fixed version.
Rescana guidance beyond those statements:
- Preserve evidence first on machines that ran it. Before you wipe anything, take copies of logs, the working directory, the
node_modulesfolder and any temporary-directory marker files, so you can answer later what was exposed. Do not rotate from the affected machine, as GitHub says. - Rotate from a clean machine, highest privilege first. Start with cloud and CI/CD credentials, npm and GitHub publishing tokens, database passwords, API keys and signing keys, then anything else in the process environment or in the files listed above. Rotate whatever a CI job could read if the job imported the library or started the server. Remember that
.envandormconfig.jsonfiles often hold database credentials. - Log the WhatsApp sessions out. Unlink the affected sessions from the WhatsApp account's linked devices list and delete the stored session-token files so that the application has to pair again. Confirm the steps in WhatsApp's own help for your account type. Treat anything a stolen session could have read or sent as exposed, including customer messages. OSV's Amazon Inspector finding says the session files permit "takeover of the installer's WhatsApp account"; the rest is Rescana's reading of the sources' "log sessions out".
- Rebuild, do not clean in place. GitHub says removal does not guarantee the machine is clean. Rebuild hosts and containers that ran 2.4.1 or server 2.10.82 from a known good image, redeploy from a lockfile that does not name them, and remove 2.4.1 and 2.10.82 from every internal registry, proxy and cache.
- Pin exact versions and verify. Pin the library and the server to exact versions you have reviewed, commit the lockfile with its integrity hashes, install with
npm ci, and review changes to those hashes. Where the library arrives through @wppconnect/server, pin it withoverrides(npm, pnpm) orresolutions(yarn). The maintainers say the 2.4.0 and 2.10.81 artifacts do not contain the reported markers, and no source we opened calls later versions reviewed, so decide on 2.4.6 and later releases yourself (Rescana's reading). - Do not rely on turning off install scripts. OSV places the trigger in the import path, so installing with scripts disabled would not have prevented it. Network egress controls, secrets that are not in environment variables, and short-lived credentials are better layers (Rescana's reading).
- Consider limiting outbound access from servers and CI runners to the messaging and lookup services they actually need, which would have blocked this upload on hosts with no reason to reach Telegram (Rescana guidance; test before enforcing).
- Record what you did: which machines, which versions, which credentials were rotated and when, and which sessions were logged out.
We will update this advisory if GitHub changes its entry or lists a patched version, if the maintainers publish a post-incident statement, or if more detail on the cause, the data received or the Telegram destination is published.
Indicators of Compromise
No source published a file hash, domain or address that proves a machine was attacked. The values below are the identifiers that OSV and the project issue #2968 publish for this package. They help you find the component and the destination. Finding the component or a marker file does not by itself show that data left your network. Rescana has added none. The Telegram bot's secret token is not in any source we used, and we do not publish it.
| Indicator | Value | Source |
|---|---|---|
| Telegram bot ID | 9034051102 |
OSV MAL-2026-20044 and MAL-2026-20043; project issue #2968 |
| Telegram chat ID that receives the upload | 7343578055 |
OSV MAL-2026-20044 and MAL-2026-20043; project issue #2968 |
| Destination path stem of the HTTPS request | api.telegram.org/bot9034051102 (the bot token follows in the real URL) |
OSV MAL-2026-20044 |
| Library file, run on import | dist/utils/diagnostics.js, loaded from dist/controllers/browser.js |
OSV MAL-2026-20044 |
| Server file, run on start | dist/util/diagnostics.js, loaded from dist/server.js |
OSV MAL-2026-20043 |
| Marker file, library | .wpp-support-diag in the temporary directory (OSV: /tmp/.wpp-support-diag) |
OSV MAL-2026-20044; project issue #2968 |
| Marker file, server | .wpp-diag in the temporary directory (OSV: /tmp/.wpp-diag) |
OSV MAL-2026-20043; project issue #2968 |
SHA-256 of dist/utils/diagnostics.js in @wppconnect-team/wppconnect 2.4.1 |
06158b832fb2aeb5482202c944bc8f23d59f5cf35c0dc66e926c2e395e5673a0 |
OSV MAL-2026-20044 |
SHA-1 of the 2.4.1 package archive (wppconnect-2.4.1.tgz) |
755274304012c7489e85c23da3c2423d184d113b |
OSV MAL-2026-20044 |
| SHA-512 integrity string of the 2.4.1 package archive | sha512-b7ifN9DEUR//0ImJojfwlbkLJ3uBLVhY4CdXBeuTE9Ht5NPvtrE/MtJioT4rDerKGBnYryf+Sy6ebFw0bWPMfg== |
OSV MAL-2026-20044 |
SHA-256 of dist/util/diagnostics.js in @wppconnect/server 2.10.82 |
a8fd315c910cf0d0e84e7a642c0fd70b8e4ca20673ba0cba1f8b69392c944f2d |
OSV MAL-2026-20043 |
SHA-1 of the server 2.10.82 archive (server-2.10.82.tgz) |
68925177e1fb4f05e1958af8ac81d4a0b22003fb |
OSV MAL-2026-20043 |
| SHA-512 integrity string of the server 2.10.82 archive | sha512-zDf9tuP7v92dvkBxwLQJeIyG8qTiRcAL1YkRzo9HEKEPr2kQB4WMEwYs4kVjJaH10KcS9OtQLrUb5pqLCCk5QQ== |
OSV MAL-2026-20043 |
Rescana's check of these values: the registry still serves the server 2.10.82 archive, and its SHA-1, its SHA-512 integrity string and the SHA-256 of its diagnostics file all match OSV's values. The registry no longer serves 2.4.1, so we could not check the three library values and show them as OSV's. The npm registry's own metadata for the server version carries the same SHA-1 and integrity string.
Two services in this behaviour are legitimate: api.ipify.org (a public IP lookup) and api.telegram.org (the Telegram Bot API). Do not block or alert on either by itself. The useful signal is the pair from a host that never used them before, and the bot and chat IDs above if your tools see request paths.
Signs worth checking beyond these values (not published indicators, Rescana guidance): new or changed credentials, API keys or database sessions that were used from unfamiliar addresses after November 15; unexpected logins or message activity on the linked WhatsApp account; and, in your logs, outbound HTTPS to Telegram from servers or runners that do not use it. We will update this advisory if OSV, GitHub or the maintainers publish more.
Third-Party Risk (TPRM) Angle
The package describes itself as a library for exporting WhatsApp Web functions to Node.js for customer service, media sending and similar automation. That puts it inside chatbots, CRM add-ons, notification services and agency-built tools, often run by a supplier on your behalf with a customer's WhatsApp account and the customer's databases and cloud keys in the same environment (Rescana's reading). The reported behaviour sends exactly those items. Rescana has no data on how many suppliers use the package or on whether any data was received.
The suppliers to ask first (Rescana guidance):
- Software vendors and agencies that built WhatsApp chatbots, messaging bridges or customer-service automation on Node.js for you.
- Managed service and hosting providers that run such applications or their CI pipelines.
- Suppliers that manage your WhatsApp Business or customer messaging channel through a self-hosted library rather than the official cloud API.
Send these five questions (Rescana guidance):
- Do any systems that serve or store our data use
@wppconnect-team/wppconnector@wppconnect/server, directly or through another package? - If so, did any build, install or running system resolve library 2.4.1 or server 2.10.82, including through a mirror, cache or old container image? On which dates?
- Which credentials, databases, cloud accounts and WhatsApp sessions could those systems read, and which were rotated, from a clean machine, and when?
- Were the affected hosts rebuilt or cleaned in place, and were evidence and logs preserved first?
- What did you find in egress logs for contact with Telegram's bot interface, and how far back do your logs go?
Ask for evidence, not reassurance: the search method and date, the lockfile or image findings, the rotation list with dates, the log period reviewed and the result. Because the maintainers say the destination had not been confirmed revoked when they last replied, "we removed the package" is not an answer to question 3.
What to do internally (Rescana guidance): inventory the two versions today, work out which machines ran them, preserve evidence, rotate from a clean machine, log the WhatsApp sessions out, rebuild the machines, pin exact versions, and chase the suppliers above. We will update this advisory if GitHub changes its entry or lists a patched version, if the maintainers publish a post-incident statement, or if more detail on the cause, the destination or the data received is published.
Need help working out which suppliers run Node.js messaging automation with access to your data? Our team can help you prioritize them and track their answers. Talk to an expert
Sources
- GitHub Advisory Database: GHSA-cxc9-87g6-qrj9 (page and API): title, type malware, severity critical, reviewed, CWE-506, no CVE, vulnerable range "= 2.4.1", no patched version, description quoted above; published 08:46 EST, November 21, 2026; checked 01:05 EST, November 22, 2026.
- OSV: MAL-2026-20044 (read through the OSV API record; source record in the OpenSSF malicious-packages repository): behaviour description, Amazon Inspector finding, indicators, file hashes and archive hashes for @wppconnect-team/wppconnect 2.4.1; credits Amazon Inspector and Bitbison; published 23:10 EST, November 15, 2026, last modified 09:30 EST, November 21, 2026; checked 01:05 EST, November 22, 2026.
- OSV: MAL-2026-20043 (read through the OSV API record): same for @wppconnect/server 2.10.82; checked 01:06 EST, November 22, 2026.
- npm registry metadata for @wppconnect-team/wppconnect: version list, publish times (2.4.1 at 18:31 EST, November 15; 2.4.6 at 08:37 EST, November 18),
latesttag, maintainers; the 2.4.1 version URL and archive return not found; checked 01:05 to 01:09 EST, November 22, 2026. Archives for 2.4.0 and 2.4.6 were downloaded and hashed; nothing was run. - npm registry metadata for @wppconnect/server: version 2.10.82 published 22:09 EST, November 15, 2026, deprecation notice quoted above, dependency range; its archive was downloaded, hashed and not run; checked 01:06 to 01:10 EST, November 22, 2026.
- npm downloads API: last week for @wppconnect-team/wppconnect: 9,737 downloads, all versions, for November 15 to 21, 2026; checked 01:05 EST, November 22, 2026.
- wppconnect-team/wppconnect, issue #2968 and its comments: the reporter's description, the 03:23 EST and 09:44 EST replies, and the maintainer's statement; checked 01:06 EST, November 22, 2026.
- wppconnect-team/wppconnect-server, issue #2734 and pull request #2733: the server-side report, removal and maintainer reply; checked 01:08 EST, November 22, 2026.
- wppconnect-team/wppconnect, pull request #2967 and commit pages (06caa8c, 27ed8e3, dee571e): removal, commit timestamps (not push times) and signature status; checked 01:06 EST, November 22, 2026.
- Rescana blog feed and sitemap: no live Rescana post mentions this package; checked 01:05 EST, November 22, 2026.



