GitHub Advisory PyPI CWE-506 sweep - 1-package 2026-08-03 batch (`instalogin1234@0.0.1` Instagram credential-harvester with Discord exfil, campaign `2026-08-instalogin1234`)

What happened

On 2026-08-03, the GitHub Advisory Database published 1 new PyPI CWE-506 (Embedded Malicious Code) advisory: instalogin1234@0.0.1. Small single-package batch relative to the 2-package 2026-08-02 batch (trongriden + wacve-utils) and the 7-package 2026-08-01 batch, but the specific-behaviour prose (Instagram credential capture + Discord exfil + browser-redirect-to-legit-site social engineering) is qualitatively distinct from the generic CWE-506 boilerplate applied to most drops.

Cluster A - instalogin1234@0.0.1 Instagram credential-harvester with Discord exfil (1 package, 2026-08-03)

| Package | Version | Notes | |---|---|---| | instalogin1234 | 0.0.1 | Instagram credential capture, Discord exfil, browser-redirect-to-legit-site social engineering |

GHSA-7929-ff6q-qmmh records the tarball masquerades as an Instagram command-line interface tool. Full behaviour chain per the advisory:

1. Masquerade: package presents itself as an Instagram CLI helper (pip install instalogin1234 → instalogin1234 CLI or import).
2. Credential capture: on execution the CLI prompts the victim for their Instagram username and password and captures both.
3. Discord exfil: captured credentials are exfiltrated to a Discord channel (attacker-controlled webhook or bot channel - a widely-abused persistent-exfil channel for Python malware families that avoids needing dedicated attacker infrastructure).
4. Social-engineering cover: to hide the theft, the package opens the legitimate Instagram website (instagram.com) in the victim's default browser, creating a plausible user experience where the victim assumes "the CLI worked, my login is normal" and does not investigate.

This social-engineering cover is the key differentiator from bare credential-capture malware - the victim gets a "successful" outcome (Instagram opens, they can log in normally in the browser), which delays detection past the immediate install-and-run window and gives the operator time to use the captured credentials before the victim notices.

Attributed by the OpenSSF malicious-packages repository to campaign 2026-08-instalogin1234. Discovery credited to security researcher kam193 with hash identifier f6ed64b38b3e872668e1d36a02c53136da1ab70ec9dacd2ac3b7d38c31794ebe.

Target profile

The instalogin1234 name shape is not a typosquat of any real Instagram library - it reads as a novice / test-account name - so the operator is likely relying on casual pip-search discovery by hobbyist / novice Python developers rather than typosquat-driven mistake-installs. Targets fit the profile of:

• Social-media automation hobbyists building follower-count bots, auto-poster scripts, or scraping tools
• Novice developers following tutorials that reference "Instagram login" packages without verifying which package is legitimate (the legitimate instagrapi or instaloader libraries are the well-known real names)
• CTF / red-team players testing Instagram-related tooling in isolated environments

Any of these categories may install the package on a host that also holds real Instagram credentials for a personal or professional account - that's where the credential capture pays off.

Registry state

Package security-yanked from PyPI during the 2026-08-03 takedown. Original version tarball no longer resolvable on the public index, but private mirrors (devpi, Artifactory, Nexus, AWS CodeArtifact, Google Artifact Registry) that cached the tarball BEFORE the takedown WILL keep serving the original version.

Related tracked activity

• The 2026-08-instalogin1234 campaign name is currently a single-package designation but the OpenSSF naming convention typically indicates additional related packages will surface if the operator continues - watch subsequent daily PyPI GHSA batches for related insta* / login* / insta-login-* variants under the same campaign name.
• Broader pattern of social-media-CLI-masquerade credential harvesters on PyPI matches earlier PyPI advisories targeting Discord, Roblox, and Fortnite token stealers - same operator profile (hobbyist-developer targeting, Discord webhook exfil, social-engineering cover for the credential capture).
• Not part of the 2025-04-tronix PyPI campaign (see pypi-2026-08-02-ghsa-malware-sweep Cluster A) - different operator, different target (Instagram accounts rather than Tron wallets), different exfil channel (Discord rather than plain HTTP POST).
• No threatActor field is set - the OpenSSF campaign name 2026-08-instalogin1234 is a campaign designation, not a named actor.

Affected packages (1)

Impact

• Any host that installed instalogin1234@0.0.1 from PyPI should treat every Instagram credential typed on that host as compromised, plus every credential the affected process had access to. GHSA-7929-ff6q-qmmh uses the CWE-506 boilerplate ("any computer that has this package installed or running should be considered fully compromised - rotate all secrets from a different computer") with additional specific behaviour prose confirming credential capture and Discord exfiltration
• Cluster A - instalogin1234@0.0.1 Instagram credential-harvester with Discord exfil (1 package, = 0.0.1, 2026-08-03): instalogin1234. GHSA-7929-ff6q-qmmh records the tarball masquerades as an Instagram command-line interface tool. On execution the package captures user login credentials - Instagram username + password - and exfiltrates them to a Discord channel (attacker-controlled webhook or bot channel, a widely-abused persistent-exfil channel for Python malware families). To hide the theft the package then opens the legitimate Instagram website in the victim's default browser, creating a plausible user experience where the victim assumes the CLI worked and their login is normal
• Attributed by the OpenSSF malicious-packages repository to campaign 2026-08-instalogin1234 - the campaign name is currently a single-package designation but the OpenSSF naming convention typically indicates additional related packages will surface if the operator continues. Discovery credited to security researcher kam193
• Package is unlikely to catch a random developer install - the name instalogin1234 is a plausible novice / test-account name shape rather than a typosquat of any real Instagram library - so the operator is likely relying on casual pip-search discovery by hobbyist / novice Python developers who want a "quick Instagram login" utility (browser-automation tutorials, social-media-scraping scripts, follower-count bots) and install whatever surfaces without carefully verifying provenance
• Package no longer retains original tarball on PyPI - yanked during the 2026-08-03 takedown. Private mirrors (devpi, Artifactory, Nexus, AWS CodeArtifact, Google Artifact Registry) that cached tarballs BEFORE the takedown WILL keep serving the original version

What to do

1. 1Grep every lockfile (requirements.txt, Pipfile.lock, poetry.lock, uv.lock, pdm.lock, conda-lock.yml) for instalogin1234. Any match is a supply-chain incident and - because the payload explicitly captures Instagram credentials - every Instagram account whose credentials were typed into the CLI on that host should be treated as compromised. Change the Instagram password, revoke every third-party app authorisation in Instagram security settings, and force-logout every session
2. 2Rotate every credential the affected Python process could reach (environment secrets, cloud API keys, SSH keys) - the CWE-506 boilerplate applies in addition to the specific Instagram credential-capture behaviour. Re-image the host if it was used for sensitive account access
3. 3Audit outbound network traffic from the affected host for Discord webhook URLs (discord.com/api/webhooks/*) and Discord API calls in the 72h window around the install - that is the confirmed exfil channel and network logs may reveal the specific attacker-controlled webhook
4. 4For all PyPI installs in CI, prefer pip install --require-hashes with a fully-hashed requirements.txt or Poetry / uv / pdm lockfiles that pin resolved hashes - hash-pinning defeats swap-of-resolved-package with a same-name malicious package
5. 5Verify instalogin1234 does not still resolve via your private mirror - internal caches routinely keep serving yanked tarballs after the public takedown

References

• GHSA-7929-ff6q-qmmh - instalogin1234 malware advisory (Instagram credential harvester, Discord exfil)

  github.com
  GitHub
• GitHub Advisory Database - recent PyPI malware advisories

  github.com
  GitHub