Vulnerability · Published Sep 2, 2026 · Assigned by VulnCheck

CVE-2026-84809Tencent AI-Infra-Guard skill-scan Analysis Bypass via Excluded Python Bytecode

Medium 6.5

In plain English

A medium-severity security flaw in AI-Infra-Guard from Tencent. Proof-of-concept exploit code is public. The chance it is exploited in the next 30 days is estimated at 0.6% (higher than 47% of all known vulnerabilities).

Severity (CVSS 3.1)6.5Medium

Exploit odds (30 days)0.6%top 52.5% of all CVEs

Exploited in the wildNo reportsCISA SSVC: poc

Attack toolingNoneno public tooling known

Official description

Tencent AI-Infra-Guard's skill-scan component excludes compiled Python bytecode files from analysis by hardcoding __pycache__ directories and .pyc/.pyo/.pyd extensions into skip lists across multiple scanning surfaces. Attackers can distribute skills with benign Python source files alongside malicious compiled bytecode that executes on import while the scanner reports a safe verdict, enabling code execution when operators install the skill.

Affected products (as reported)

Vendor	Product	Versions
Tencent	AI-Infra-Guard	0 – before 4.5.24.6.0 – through 4.6.0
Tencent	aig-skill-scan	0 – through 0.2.1

Weakness: CWE-693 · Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:N/I:H/A:N

References

GitHub Issue #531 issue-tracking
https://github.com/Tencent/AI-Infra-Guard/commit/7e0f749e3c023e5c6ab7b32fe97b3f6f2e8aeb04 patch
https://github.com/Tencent/AI-Infra-Guard/blob/v4.6.0/skill-scan/skill_scan/tools/dir/dir_actions.py technical-description
https://github.com/Tencent/AI-Infra-Guard/blob/v4.6.0/skill-scan/skill_scan/utils/pre_scan.py technical-description
https://github.com/Tencent/AI-Infra-Guard product
VulnCheck Advisory: Tencent AI-Infra-Guard skill-scan Analysis Bypass via Excluded Python Bytecode third-party-advisory