Back to skill

Security audit

data-researcher

Security checks for vulnerabilities and agentic risk

Overview

This is a plain Markdown data-research assistant skill with broad but disclosed research capabilities and no embedded code, persistence, or hidden execution behavior.

Install only where the agent's data access is already controlled. For private databases, logs, scraping targets, or internal sources, make sure the user has authorization and gives clear task boundaries before the skill is used.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill’s invocation and execution scope are broadly defined, with no clear trigger boundaries or permission constraints before it begins querying context and assessing data requirements. In an agentic environment, this can cause the skill to activate for loosely related prompts and proceed into data discovery or collection behaviors that may exceed user intent or organizational policy.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly includes sensitive-capability areas such as database access, web scraping, private sources, real-time streams, and log analysis, but provides no privacy warning, authorization requirement, or restriction on collecting protected data. In practice, this can normalize unauthorized collection or analysis of personal, confidential, or internal data, increasing the risk of privacy violations, policy breaches, and improper data exfiltration.

Content

No source excerpt is available for this finding.