lgpd-data-mapping

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local bash script (assets/extract-prisma-annotations.sh) to parse Prisma schema files. The script is restricted to extracting comments prefixed with @lgpd: using awk.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface (Category 8) as it ingests and processes untrusted data from local project files to build reports.\n
  • Ingestion points: Reads prisma/schema.prisma, API route definitions, third-party SDK configurations, and system log files.\n
  • Boundary markers: Absent. Content is read into the agent's context without explicit isolation delimiters or specific instructions to ignore embedded instructions in the data.\n
  • Capability inventory: The skill can read/write to the file system and execute shell scripts.\n
  • Sanitization: Absent. The parsing script and agent analyze the content without validation or filtering for adversarial prompt patterns that might be hidden in code comments or logs.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 02:52 PM
Security Audit — agent-trust-hub — lgpd-data-mapping