audit-context-building

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface inherent to its purpose as a code analysis tool.\n
  • Ingestion points: Data enters the agent context through the <target-path> provided for analysis, which may contain untrusted code, as described in Phase 1 (Initial Reconnaissance) and Phase 2 (Deep Analysis).\n
  • Boundary markers: There are no specific delimiters or instructions defined in the skill to isolate the content of the analyzed files or to warn the agent against executing instructions found within them.\n
  • Capability inventory: The skill utilizes powerful tools including Read, Glob, Grep, Bash, and Write as specified in the SKILL.md toolset.\n
  • Sanitization: The skill does not implement sanitization, escaping, or validation of the content of the analyzed source code before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 08:44 AM