read-rss-aggregator
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill fetches and processes content from external RSS feeds, creating an indirect prompt injection attack surface. If a blog post contains text designed to look like AI instructions, the agent might inadvertently follow them.\n
- Ingestion points:
scripts/aggregate.pyretrieves content from the URLs defined inreferences/feeds.opml.\n - Boundary markers: The summaries are presented in a list format but do not include explicit instructions for the agent to ignore any command-like text found within them.\n
- Capability inventory: The script uses network access to fetch data and provides the resulting text directly to the agent's context.\n
- Sanitization: The
clean_summaryfunction inscripts/aggregate.pystrips HTML tags and truncates output, but does not identify or remove potential prompt injection patterns.\n- [EXTERNAL_DOWNLOADS]: The skill makes network requests to fetch feed data from external sources listed inreferences/feeds.opml. These sources are mostly established technical blogs and well-known platforms.
Audit Metadata