Back to skill

Security audit

analyze-patent-family-ip

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent patent-family analysis workflow with a local HTML renderer and no hidden persistence or unrelated data access found.

Before installing, confirm you are comfortable using PatSnap MCP services for the patent identifiers and evidence being analyzed. The local script appears limited to rendering a supplied JSON file into an offline HTML report, but the workflow still depends on external patent-data connectors and should not be treated as legal advice.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The declared description presents this skill as an end-to-end patent family analysis tool starting from a patent identifier or PatSnap URL and performing substantive analytical steps. The code chunk, however, is only a downstream renderer: it loads normalized JSON, validates schema consistency, escapes content, and emits an offline HTML report with embedded local JS/CSS. The docstring explicitly states it does not search patents, call an MCP server, or perform AI analysis, and that an upstream workflow creates the normalized JSON. While the report contents correspond to the same patent-family-analysis domain, the primary behavior in this code is safe presentation of already-derived results, not performing the analysis described. Therefore this is a material description/behavior mismatch.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 489)May include surrounding context.

md
scripts/analyze_family.py

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 495)May include surrounding context.

md
scripts/analyze_family.py

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding

The skill advertises network, file read, and file write behavior but declares no explicit tool/permission scope. That creates an over-privileged execution surface where an agent or runtime may permit broader actions than users expect, increasing the risk of unintended data access, outbound requests, or filesystem modification if the skill is invoked in a permissive environment.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The instruction "Use the English interface and English output" imposes a fixed language choice. This conflicts with the stated policy criterion for locale/language restrictions because the skill does not offer the user a choice or frame the constraint as an explicit, justified region-specific requirement.

Content

No source excerpt is available for this finding.