startup-business-analyst-business-case

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection (Category 8) as it is designed to ingest and analyze external materials such as pitch decks and market research data.
  • Ingestion points: The skill uses the Read tool to access user documents and WebFetch to retrieve web content (SKILL.md).
  • Boundary markers: No delimiters or instructions to ignore embedded commands are specified in the document structure.
  • Capability inventory: The skill utilizes Bash, Write, WebFetch, and Edit tools (SKILL.md).
  • Sanitization: There is no evidence of input validation or content filtering for the data being processed.
  • [COMMAND_EXECUTION]: The skill requests permission to use the Bash tool. While no malicious shell scripts are included in the definition, granting shell access to an agent that processes untrusted external content increases the potential impact of a prompt injection attack.
  • [EXTERNAL_DOWNLOADS]: The skill employs WebSearch and WebFetch to gather information from external domains. Although used here for legitimate market analysis, these capabilities could be misused for data exfiltration if the agent's instructions were compromised.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 07:42 PM