llm-prompt-injection

Warn

Audited by ZeroLeaks on Apr 16, 2026

Risk Level: MEDIUM
Scan Summary

This skill looks risky overall, with medium confidence. The skill has 3 transparency concerns that weaken pre-use reviewability, mainly around obfuscated or encoded execution path and remote script execution without review boundary. The scanned skill keeps data and instructions reasonably separate and does not strongly encourage the agent to treat external content as policy. Behavior analysis was not run.

Score
69/100
Verdict
AT_RISK
Confidence
medium
Findings
3
Section Analysis (3)
TransparencyAT_RISK
17/100

The skill has 3 transparency concerns that weaken pre-use reviewability, mainly around obfuscated or encoded execution path and remote script execution without review boundary.

Prompt InjectionPASS
92/100

The scanned skill keeps data and instructions reasonably separate and does not strongly encourage the agent to treat external content as policy.

Agent BehaviorSkipped

Behavior analysis was not run.

Findings (3)
CRITICAL

Remote script execution without review boundary

HIGH

Obfuscated or encoded execution path

HIGH

Hidden or dense payload-like content

Coverage DetailsClick to expand
Discovered Files
1
Omitted Files
0
Analyzed Bytes
12.7 KB
Sections Completed
2
Sections Skipped
1
Behavior Probes
0/0
Audit Metadata
Score
69/100
Verdict
AT_RISK
Confidence
medium
Sections
2/3 completed
Findings
3
Mode
risk
Files Scanned
1
Duration
0.1s
Analyzed
Apr 16, 2026, 09:01 AM
Security Audit — zeroleaks — llm-prompt-injection