← back to prompt-injection-defense
SkillSpector · prompt-injection-defense
independent scanner by NVIDIA · skill by seb1n · how it works ↗
Without declared permissions the skill's intent is opaque and cannot be validated.; Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, dat…; Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous …
scanned 2026-08-23
Findings (4)
Without declared permissions the skill's intent is opaque and cannot be validated.
SKILL.md
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
scripts/audit_boundary_manifest.py
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
scripts/audit_boundary_manifest.py
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).
references/defense-patterns.md
What the verdicts mean
SkillSpector reports on SkillMD's shared five-tier scale. See how SkillSpector works ↗.
Overall severity LOW (risk score in the safe range)
Overall severity MEDIUM
Overall severity HIGH
Overall severity CRITICAL
Scan could not complete