competitor-monitoring-system

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an Indirect Prompt Injection surface because it is designed to ingest and analyze data from multiple attacker-controlled external environments.
  • Ingestion points: The playbook instructs the agent to monitor and scrape content from competitor blogs, social media profiles (LinkedIn, Twitter), community forums (Reddit, Hacker News), and third-party review platforms (G2, Capterra).
  • Boundary markers: There are no specified delimiters or instructions to the agent to disregard potential commands or malicious formatting embedded within the scraped external content.
  • Capability inventory: While the skill primarily performs analysis, it is instructed to write structured intelligence reports and baseline files to the local directory structure (clients/).
  • Sanitization: The instructions do not define any validation, filtering, or sanitization steps for the data retrieved from external URLs before processing it.
  • [NO_CODE]: This skill is a 'playbook' that provides high-level instructions and orchestrates existing tools rather than delivering its own scripts or binary executables, which reduces the risk of direct malicious code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 10:48 AM