Back to skill

Security audit

Stocktorch

Security checks for vulnerabilities and agentic risk

Overview

This stock-analysis skill is purpose-aligned and mostly transparent, but users should know it performs web searches, may save reports locally, and the packaged artifact only contains instructions, not the referenced scripts or docs.

Install only if you are comfortable with a stock-research helper that may search the web and save generated analysis reports locally. Treat its investment output as research assistance, not financial advice, and note that this package appears to lack the referenced analysis scripts and documentation.

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
  • 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 (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The manifest trigger includes broad phrases such as "分析股票" and "投资建议", which can match ordinary financial conversation and cause the skill to activate outside clearly intended contexts. In this skill, unintended activation is more concerning because activation can lead to web searches, file generation, and potentially authoritative-looking investment output without explicit user intent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The market-analysis trigger example "大盘怎么样" is vague and likely to overlap with common casual chat about markets, creating a risk of accidental invocation. Because the skill then performs real-time search and generates a saved report, ambiguous activation can expose users to unexpected tool use and persistence of generated content.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill instructs the agent to save generated reports into the reports/ directory, but the user-facing description does not clearly warn that outputs will be written to disk. This matters because reports may contain user-request-derived content, searched market information, and timestamps, creating unexpected data persistence and possible privacy or workspace hygiene issues.

Content

No source excerpt is available for this finding.