Back to skill

Security audit

Elixir Dev

Security checks for vulnerabilities and agentic risk

Overview

This Elixir/Phoenix helper is documentation-only and purpose-aligned, with caution needed around destructive database commands and delete examples.

This skill appears safe to install as a development reference. Before following generated commands, review any database, migration, delete, reset, or production-environment action carefully and confirm it targets the intended local or test system.

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
Confidence
95% confidence
Finding
The manifest description says to use the skill for "any Elixir or Phoenix development task," followed by broad examples like writing modules, fixing tests, and refactoring code. This is an ambiguous, expansive activation condition that could overlap with many ordinary development requests without clearly defining boundaries or exclusions.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The LiveView example handles a "delete" event by calling `delete_item!` immediately, but the markdown provides no warning that this operation removes data or that confirmation should be considered. In markdown guidance, destructive behaviors that may affect user data should be accompanied by a warning or note about confirmation and reversibility.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The markdown reference includes `mix ecto.drop` and `mix ecto.reset`, which can delete and recreate a database, but the file provides no warning that these commands are destructive. Because this is a documentation file, the omission of a user-facing caution about potential data loss matches the missing-warning criterion.

Static analysis

No suspicious patterns detected.