Security audit

PUA

Security checks for vulnerabilities and agentic risk

Overview

This skill has no executable payload, but it broadly changes agent behavior and retry limits across all task types without clear safety boundaries.

Review this skill carefully before installing. It is not showing code execution or data theft, but it is designed to make the agent persist harder and escalate after failures, so it should only be used when you explicitly want that behavior and can supervise tool use, costs, and side effects.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

Findings (1)

T01 · Skill Instruction Hijacking

Error

Location: SKILL.md:2
Finding: Broad Trigger-Based Agent Role and Stopping-Policy Hijacking
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The trigger description is extremely broad and includes ordinary frustration phrases, repeated-failure language, and 'all task types,' which makes unintended activation likely across unrelated contexts. Because the skill explicitly aims to override normal behavior with pressure-driven exhaustive action, accidental invocation can materially alter agent conduct, increasing the risk of unsafe persistence, scope creep, and degraded user-aligned behavior.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The README-style content reinforces always-on escalation behavior without defining exclusions, safety boundaries, or cases where the skill must defer to higher-priority safety rules. In context, a skill built around 'keep going' and pressure escalation is more dangerous than a normal productivity aid because ambiguous activation could push the agent to ignore sensible stopping conditions or continue beyond intended scope.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 83% confidence
Finding: The listed variants include '/pua:pua-ja (Japanese)' and multilingual trigger phrases elsewhere, but the file does not state that language switching occurs only with explicit user choice. That creates a risk of forcing a specific language or locale behavior without opt-in, which conflicts with the language/locale policy requirement.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.