Security audit

Defense

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed informational legal-defense skill with no code or system access, but users should treat its guidance as general and consult a lawyer.

Before installing, understand that this skill provides general legal-defense information and decision-framing, not legal advice for your specific situation. In criminal, civil, or regulatory matters, confirm deadlines and strategy with a licensed attorney in the relevant jurisdiction.

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 (5)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The examples provide concrete legal-defense guidance in a high-stakes domain without clearly warning that the content is general information rather than legal advice and cannot substitute for a qualified attorney. Users may rely on these prompts and responses during criminal, civil, or regulatory matters, where jurisdiction-specific differences and case facts can materially change the correct course of action, increasing the risk of harmful decisions.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The phrase 'Run immediately when any legal threat or action begins' is very broad and does not define boundaries for what counts as a triggering legal threat. Without narrower scope, exclusion conditions, or negative examples, the skill could be invoked for everyday disputes or informal complaints that are not intended targets.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The trigger list includes broad, common legal terms such as "defense," "investigation," "subpoena," and "bail," which could cause the skill to activate in contexts where the user did not intend to invoke legal-defense guidance. Unintended activation is especially risky here because the skill addresses high-stakes legal situations, so irrelevant or premature guidance could confuse users or displace more appropriate routing.
Content: No source excerpt is available for this finding.

Vague Triggers

Low

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The instruction 'While a matter is active' is ambiguous because it does not clarify what counts as an active matter or whether the check should occur on a schedule, after events, or only in certain case types. The absence of scope limits or exclusion examples makes the trigger imprecise.
Content: No source excerpt is available for this finding.

Vague Triggers

Low

Category: Not specified by scanner
Confidence: 83% confidence
Finding: The phrase 'Before any significant decision in a defense matter' relies on a subjective threshold and does not define which decisions are in scope beyond a non-exhaustive list below. Because the trigger lacks explicit boundaries or exclusions, users may invoke it inconsistently or in unintended contexts.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.