ClawHub
Back to skill

Security audit

Judge

Security checks for vulnerabilities and agentic risk

Overview

This court-help skill is not malicious, but it should be reviewed because it gives broad, deadline-sensitive legal guidance without clear legal-advice limits or verification requirements.

Review carefully before installing. Use this only as general court-process orientation, not as legal advice. Verify every deadline, filing format, service rule, appeal period, and courtroom requirement against official court materials, clerk instructions, or a licensed attorney, especially for criminal, immigration, custody, eviction, restraining-order, appeal, or urgent deadline matters.

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

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 (4)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

This skill repeatedly prompts users to rely on the agent for deadline calculation, filing requirements, hearing preparation, discovery obligations, settlement evaluation, appeals, and enforcement, but it does not warn that the information may be incomplete, jurisdiction-specific, or not a substitute for a lawyer, clerk, or official court instructions. In a court context, users may reasonably treat these prompts as authoritative; a mistaken response could cause missed deadlines, defective filings, waived rights, or noncompliance with court orders.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list includes broad, high-frequency legal terms such as "court," "hearing," "judge," "order," and "settlement," which can match many ordinary conversations and cause the skill to activate outside clearly intended legal-navigation contexts. In a legal assistance skill, unintended activation is especially risky because it may inject procedural or quasi-legal guidance into ambiguous situations, increasing the chance of irrelevant, overconfident, or harmful advice being surfaced at the wrong time.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill description is broad and does not define clear activation boundaries, jurisdictions, or excluded use cases, which can cause the agent to invoke the skill in inappropriate legal contexts. In a legal-assistance skill, ambiguous scope increases the chance of overreaching into fact-specific legal advice or being relied on where the skill is not suitable.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

This skill provides court and procedural guidance but does not present a prominent upfront warning that it is not a lawyer, not legal advice, and may be incomplete or jurisdiction-specific. Because court deadlines, filing rules, and hearing conduct can directly affect housing, custody, finances, or liberty, users may reasonably over-rely on the output and suffer serious harm from inaccurate or incomplete guidance.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.