edr-bypass-re
Fail
Audited by Snyk on Nov 20, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (medium risk: 0.30). The skill openly advertises EDR-evasion techniques for authorized lab evaluation, so its dangerous content falls under separate rules, but it also gives the agent guidance on silencing telemetry and hitting hardened hosts that goes somewhat beyond the stated measurement purpose, which under the recall-first override makes it medium rather than no.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The required workflow fetches public community code and writeups (SysWhispers3, Hell's Gate/Halo's Gate, Ekko, pe-sieve) that the agent must actively select and pull, so outsider-authored text is exposed only through a search/fetch path, not an unsolicited feed.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (medium risk: 0.30). The only operational locator is the pe-sieve GitHub repository in the Tools table's Install column, which points to an individual developer's project (hasherezade), so it is reputable but unofficial and treated as medium, while the attribution link and other tool names are not inbound dependencies.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (medium risk: 0.65). The skill directs the agent to implement and run EDR evasion techniques (ntdll unhooking, AMSI and EtwEventWrite patching, syscall stubs) on a Windows host, which is bypassing security mechanisms on the machine it runs on, though it is framed around authorized lab use with scope boundaries.
Issues (4)
E004
CRITICALPrompt injection detected in skill instructions.
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