content-monitor

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
  • Ingestion points: External content is retrieved from multiple untrusted sources including YouTube channel RSS feeds, Reddit JSON APIs, GitHub trending pages, and general RSS/Atom feeds (documented in scripts/content-monitor/sources.json and scripts/content-monitor/channels.json).
  • Boundary markers: The skill documentation does not specify the use of delimiters or 'ignore instructions' markers when interpolating extracted content into the prompts for the 'Summarize' (MiniMax) or 'Evaluate' (GPT-5.4) phases.
  • Capability inventory: The skill utilizes Bash, Read, and Grep tools and executes several local JavaScript adapters (e.g., discover-rss.js, discover-github.js) located in scripts/content-monitor/ to manage discovery and extraction.
  • Sanitization: No explicit sanitization, filtering, or validation logic is defined to inspect raw external text before it is processed by the underlying language models.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 04:07 PM