ClawHub
Back to skill

Security audit

Stock Research

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent stock-research tool, but it has under-scoped persistent file writes to fixed local paths and a task-log instruction that users should review before installing.

Review this skill before installing. Its market-data access is purpose-aligned, but users should ensure outputs are written only to their intended workspace, disable or approve record.md logging explicitly, and treat generated stock recommendations as informational rather than investment advice.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Context-Inappropriate Capability

Low
Confidence
95% confidence
Finding
The skill instructs the agent to update a local `record.md` after task completion, which is unrelated to stock analysis and expands behavior into filesystem modification. Even though the target is only a markdown log, this creates unnecessary side effects and can normalize unauthorized local file writes beyond the user’s requested task.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The activation conditions are very broad and can match many ordinary finance-related requests such as stock analysis, recommendations, or quantitative selection. This can cause the skill to auto-activate in situations where the user did not clearly request this specific tool, increasing the chance of unintended execution of its scripts, data-fetching, and side-effect behavior.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The script writes output to a hard-coded user-local path without prompting, path validation, or making the destination configurable. In an agent/skill context, silent writes can overwrite existing files, leak sensitive analysis artifacts into predictable locations, and create persistence side effects the user did not explicitly authorize.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The generated report is also written to a fixed absolute path without prior disclosure or confirmation. In a skill environment, this creates non-transparent filesystem side effects and may expose potentially sensitive financial analysis in a predictable location accessible to other local processes or users with access to that profile.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.