ClawHub
Back to skill

Security audit

Global Market Snapshot

Security checks for vulnerabilities and agentic risk

Overview

This skill fetches public market data from disclosed finance sources to help generate a market snapshot, with no evidence of hidden persistence, credential access, exfiltration, or destructive behavior.

Install this if you are comfortable with an agent making live web requests to public finance sites and receiving raw market payloads that the agent must parse. Treat the output as informational market data, not financial advice, and expect occasional N/A values if the public sources fail or change format.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (4)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The declared behavior says the skill generates a complete structured market snapshot, but the implementation guidance only fetches raw single-ticker payloads and pushes parsing, fallback, and formatting to the caller. This mismatch is dangerous because agents may invoke the skill under false assumptions, mishandle raw outputs, or trust incomplete data paths that were never safely encapsulated by the skill.

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill instructs use of shell-based network fetching via curl and related helpers, but declares no tool scope or permission boundaries. In an agent environment, this can lead to unintended shell or network access beyond what a reviewer or orchestrator expects, increasing the risk of misuse and reducing policy enforcement.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The description says to use the skill when the user asks for 'market data', 'stock market report', or 'daily market summary'. These phrases are fairly broad and there are no negative examples or tighter scope constraints, so the skill could be invoked for loosely related finance questions rather than only for this specific snapshot workflow.

Natural-Language Policy Violations

Low
Confidence
88% confidence
Finding
The skill instructs the caller to generate output with both a UTC timestamp and 'China time'. This imposes a specific locale on all users, and the file does not offer a language/locale choice or explain why China time is always required.

Static analysis

No suspicious patterns detected.