ClawHub
Back to skill

Security audit

Json Parser

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed construction JSON parsing helper with minor scope drift but no evidence of hidden, destructive, or persistent behavior.

Before installing, treat this as a file-processing helper: provide only files you intend the agent to read, and be aware that the runtime instructions mention CSV/Excel handling even though the documented code is JSON-focused.

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 (2)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The instructions expand the skill from JSON parsing into broader CSV/Excel input handling and multi-format export behavior, which increases the operational scope beyond the declared skill purpose. This can cause downstream agents to process unexpected file types or perform unintended data-handling actions, weakening boundary enforcement and making misuse or unsafe file-processing paths more likely.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

The file frames the capability as a general construction project management assistant rather than a narrowly scoped JSON parser. While not overtly malicious, this scope drift can encourage use outside intended boundaries, leading the agent to answer broader operational requests or invoke unrelated logic without clear authorization or validation expectations.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.