Malicious code in deepface-weights (PyPI)
The PyPI package 'deepface-weights' versions 0.1.0, 0.1.1, and 0.1.2 contains malicious code that exfiltrates Telegram session files. Upon import, it starts a daemon thread that polls for a Telegram session file in the current directory and sends it along with the local username to a hardcoded attacker-controlled endpoint. This allows attackers to fully take over the associated Telegram account. The package masquerades as a face-recognition library but contains no legitimate functionality.
AI Analysis
Technical Summary
The 'deepface-weights' PyPI package (versions 0.1.0, 0.1.1, 0.1.2) contains embedded malicious code that, upon import, initiates a background thread polling every 10 seconds for the presence of a Telegram session file named 'data/telethon_market_userbot.session' in the current working directory. When found, it exfiltrates this file along with the local OS username to a hardcoded webhook URL controlled by the attacker. Since the Telegram session file contains authenticated credentials, this exfiltration enables full compromise of the victim's Telegram account. The package metadata is clearly fake and the package does not provide any legitimate deepface functionality. Source code comments in Russian confirm the destination is the attacker's server.
Potential Impact
Successful exploitation results in the theft of Telegram session credentials, allowing attackers full control over the victim's Telegram account. This can lead to unauthorized access, message interception, impersonation, and further compromise of the victim's contacts and data within Telegram.
Mitigation Recommendations
Users should immediately uninstall the 'deepface-weights' package versions 0.1.0, 0.1.1, and 0.1.2 and avoid installing it. Since no official patch or fix is available, the best mitigation is to remove the malicious package and revoke any Telegram sessions that may have been compromised. Users should verify their Telegram account security and reset credentials if suspicious activity is detected.
Malicious code in deepface-weights (PyPI)
Description
The PyPI package 'deepface-weights' versions 0.1.0, 0.1.1, and 0.1.2 contains malicious code that exfiltrates Telegram session files. Upon import, it starts a daemon thread that polls for a Telegram session file in the current directory and sends it along with the local username to a hardcoded attacker-controlled endpoint. This allows attackers to fully take over the associated Telegram account. The package masquerades as a face-recognition library but contains no legitimate functionality.
Affected software
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
The 'deepface-weights' PyPI package (versions 0.1.0, 0.1.1, 0.1.2) contains embedded malicious code that, upon import, initiates a background thread polling every 10 seconds for the presence of a Telegram session file named 'data/telethon_market_userbot.session' in the current working directory. When found, it exfiltrates this file along with the local OS username to a hardcoded webhook URL controlled by the attacker. Since the Telegram session file contains authenticated credentials, this exfiltration enables full compromise of the victim's Telegram account. The package metadata is clearly fake and the package does not provide any legitimate deepface functionality. Source code comments in Russian confirm the destination is the attacker's server.
Potential Impact
Successful exploitation results in the theft of Telegram session credentials, allowing attackers full control over the victim's Telegram account. This can lead to unauthorized access, message interception, impersonation, and further compromise of the victim's contacts and data within Telegram.
Defensive Guidance
Users should immediately uninstall the 'deepface-weights' package versions 0.1.0, 0.1.1, and 0.1.2 and avoid installing it. Since no official patch or fix is available, the best mitigation is to remove the malicious package and revoke any Telegram sessions that may have been compromised. Users should verify their Telegram account security and reset credentials if suspicious activity is detected.
Technical Details
- Gcve Source
- db.gcve.eu
- Osv Id
- MAL-2026-14132
- Osv Schema Version
- 1.7.4
- Aliases
- []
- Ecosystems
- ["PyPI"]
- Database Specific Severity
- null
- Cvss Version
- null
Threat ID: 6a84fa65c6e8be0332ef1307
Added to database: 08/19/2026, 00:35:49 UTC
Last enriched: 08/19/2026, 00:50:58 UTC
Last updated: 08/19/2026, 01:04:29 UTC
Views: 3
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
External Links
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.