competitor-monitor

Warn

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the execution of a local Python script competitor-tracker.py (Step 3). The source code for this script is not provided within the skill package, making its behavior and safety unverified.
  • [COMMAND_EXECUTION]: The skill accesses the file system to read brand profiles, guidelines, and SOPs from the user's home directory (~/.claude-marketing/). While these paths are skill-specific, accessing data in the home directory should be monitored.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from competitor websites.
  • Ingestion points: Website meta tags, page titles, H1 headings, and core messaging blocks captured from external competitor URLs (Step 2).
  • Boundary markers: Absent. There are no instructions to the agent to distinguish between its own logic and instructions that might be embedded in the crawled content.
  • Capability inventory: The agent has the ability to execute the competitor-tracker.py script and read local configuration files.
  • Sanitization: Absent. No validation or sanitization is performed on the data retrieved from external sources before it is processed or saved.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 27, 2026, 01:02 PM