bbb-estatisticas

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection vulnerability surface. The agent is instructed to process content from Wikipedia and the broader web to answer user queries. \n
  • Ingestion points: External data enters the agent's context through local scraping scripts (scripts/bbb-geral.py, scripts/bbb-historico.py, scripts/bbb-participantes.py) and the search_web tool results. \n
  • Boundary markers: The skill lacks explicit instructions or markers to delimit external content from system instructions, which could allow maliciously crafted external data to influence agent behavior. \n
  • Capability inventory: The skill has access to network operations (via scraping scripts) and a web search tool. \n
  • Sanitization: There is no evidence of sanitization or filtering applied to the retrieved text before it is presented to the LLM. \n- [EXTERNAL_DOWNLOADS]: The Python scripts in the scripts/ directory perform HTTP GET requests to pt.wikipedia.org to retrieve information. This is a well-known and trusted source for general information.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 10:08 PM