hypervisor-introspection
Warn
Audited by Snyk on Nov 14, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The required workflow is mostly local CLI enumeration and lab reproduction, but the payloads guidance tells the agent to search GitHub for a public VENOM PoC, which means ingesting arbitrary outsider-authored repositories it must actively select (low, not medium, since no outsider text is pushed into a monitored queue).
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (medium risk: 0.30). The decisive boundary is the instructed setup path that clones and builds code from an individual-owned repository (tandasat/HyperPlatform, which is then loaded as a kernel driver), so this is a common/reputable unofficial dependency and the highest tier is medium, while the QEMU, PANDA, and Bitdefender KVMi clones are strong/official.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (medium risk: 0.40). The skill is a security-knowledge reference, but it tells the agent to install kernel-mode drivers and hypervisors on a host, write system config under /etc, and stop system services, which are machine-state changes that can require elevated privileges.
Issues (3)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata