Security audit

akylegal-applications

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent legal drafting guide, but it needs review because it gives PRC-specific legal procedure and asks for sensitive personal or financial details without enough scoping or privacy guidance.

Before installing, treat this as a PRC-focused legal drafting aid, not general legal advice. Confirm the jurisdiction and current law before relying on deadlines or forms, and use placeholders or redacted values for ID numbers, addresses, phone numbers, bank accounts, property records, and other sensitive data unless a lawyer confirms they are necessary.

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

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (11)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger description is extremely broad, covering many common legal-related terms such as lawsuit, legal documents, strategy, appeal, and procedural guidance. This can cause the skill to activate in situations where users only need general information, increasing the chance of unnecessary skill invocation, overcollection of sensitive legal facts, or generation of high-stakes procedural guidance outside the intended scope.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The general formatting rules instruct inclusion of highly sensitive personal data such as full ID numbers, date of birth, address, and phone number in a broad reusable template. In a general-purpose AI skill, this can normalize over-collection and cause users to disclose more personal information than is necessary for drafting assistance, increasing privacy and data-handling risk.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The template collects extensive personal identifiers but provides no warning about privacy, retention, redaction, or lawful handling. In an AI-assisted workflow, that omission can lead users to enter real third-party or client PII directly into prompts or stored documents without understanding the sensitivity.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The templates cite PRC-specific statutes and procedural rules as if generally applicable, without prominently limiting the material to PRC jurisdiction. Because this is a broad legal-applications skill, users may wrongly rely on these forms and deadlines in other jurisdictions, creating legal error, missed deadlines, or invalid filings.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The enforcement application template asks for detailed asset leads such as bank accounts, property certificates, vehicle details, and other assets, but gives no warning that this information is highly sensitive and must be obtained and used lawfully. In the context of an AI skill, this could facilitate unsafe disclosure of financial data or misuse of third-party asset information.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 87% confidence
Finding: The guide gives procedural and strategic legal advice using terminology and examples that appear specific to a Chinese legal context, but it does not clearly disclose jurisdictional scope. Users could mistakenly apply these instructions in another country or court system, leading to harmful procedural errors, invalid filings, waived rights, or flawed trial conduct.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The document immediately presents arbitration rules, institutions, deadlines, and statutes under PRC law without an upfront, prominent statement that the guidance is China-specific. In a legal-procedure skill, users may reasonably rely on this as generally applicable arbitration advice, creating a material risk of misfiling, missed deadlines, invalid clauses, or incorrect procedural strategy in other jurisdictions.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 86% confidence
Finding: The guide is primarily written in English, but line L067 includes untranslated Chinese text ("正当理由"). This creates a language consistency issue and effectively imposes mixed-language content on users without opt-in or explanation, which matches the policy concern about forced language/locale usage.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 88% confidence
Finding: This markdown file contains natural-language content that assumes a specific legal locale and partially forces Chinese-language terminology, such as Chinese statute citations and untranslated terms. Under the policy, language or locale constraints should be explicitly offered as a choice or clearly justified; this file does not provide such opt-in framing.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The phrase "unexplained巨额财产来源不明" embeds Chinese text directly into an English sentence without translation or user choice. This is a language-policy concern because it forces a specific language context rather than presenting a localized or opt-in alternative.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The sentence ends with the Chinese phrase "正当理由" inside otherwise English prose, which creates an implicit language requirement. The policy allows locale constraints when documented and justified, but no such explanation appears in the document.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.