ClawHub
Back to skill

Security audit

polymarket-predictradar-news-impact-skills

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed, read-only Polymarket news analysis workflow with some routing and accuracy caveats but no hidden execution, persistence, or mutation behavior.

Install this if you want Polymarket news-impact analysis, but treat outputs as informational rather than financial advice. Be aware it may activate on broad news phrasing, and verify non-English entity translations and any external data-layer skill before relying on results.

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 (2)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger phrases are broad enough to match ordinary requests about news or market reactions, which can cause the skill to activate outside its intended scope. In an agent system, overscoped triggering can route users away from more appropriate skills, causing unintended web searches, market analysis, or disclosure of irrelevant financial-style outputs based on ambiguous prompts.

Natural-Language Policy Violations

Medium
Confidence
89% confidence
Finding
Forcing English-only query behavior for Chinese news without user choice can lead to inaccurate entity translation, missed markets, or misleading results while the user believes they are getting faithful coverage of the original news. This is primarily an integrity and UX safety issue rather than a direct security exploit, but in a market-analysis context it can still influence user decisions with incomplete or distorted outputs.

Static analysis

No suspicious patterns detected.