Security audit

Legal Regulatory Compliance Mapper

Security checks for vulnerabilities and agentic risk

Overview

This is a document-only legal compliance helper, but some examples push uncited, jurisdiction-specific legal requirements and deadlines despite the skill's own verification limits.

Review or revise the example scenarios before installation. The skill is not executing code, but users may over-rely on its sample legal outputs; require authoritative sources and qualified counsel for every jurisdiction-specific obligation, filing deadline, license, regulator, and timeline.

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

MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (6)

Intent-Code Divergence

High

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill explicitly says not to invent citations, deadlines, regulator names, thresholds, or license requirements unless backed by authoritative sources, yet its examples present concrete legal obligations and deadlines as if they are settled facts. In a legal/compliance workflow, users may rely on these examples as templates for real-world advice, causing the model to emit unsupported jurisdiction-specific claims that could lead to regulatory noncompliance or unauthorized legal guidance.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

High

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The acceptance criteria and safety section say the skill should generate checklists and planning aids rather than legal conclusions, but the usage scenarios direct the agent to produce concrete compliance determinations, filing deadlines, licensing steps, and jurisdiction-specific instructions. This contradiction weakens safety boundaries and can steer downstream outputs into actionable legal advice, increasing the chance of hallucinated or outdated regulatory guidance being treated as authoritative.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The acceptance criteria explicitly require that SKILL.md and README.md are "English-first," which imposes a language preference as a policy requirement. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation unless the constraint is justified or optional.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The package rule states that the package must remain "English-first," again enforcing a locale/language constraint in natural language. Because no user choice, opt-in, or region-specific justification is provided, this conflicts with the language/locale policy criteria.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 87% confidence
Finding: This markdown file contains a full scenario and expected output in Chinese, while the manifest header declares language: en and the acceptance criteria state the documentation is English-first. Under the policy rule, forcing or embedding a specific language without user opt-in can be a locale/language policy issue when no choice or justification is provided.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The manifest sets the skill language to "en" with no indication that users can choose another language or that English is required for a documented region-specific reason. This is a natural-language policy concern because it imposes a locale constraint without opt-in.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.