skills/moshe-ship/hurmoz/majal/Gen Agent Trust Hub

majal

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external JSONL datasets which may contain untrusted content. This creates a surface for indirect prompt injection where malicious instructions embedded in the data records could attempt to influence the agent's behavior while it is scanning or sampling the file.
  • Ingestion points: The skill instructions guide the agent to use majal scan, majal stats, and majal sample on user-provided .jsonl files.
  • Boundary markers: The instructions do not define specific delimiters or "ignore instructions" wrappers for the data being processed.
  • Capability inventory: The skill relies on the majal CLI tool for file reading and data analysis. It also supports file writing via the majal fix --output command.
  • Sanitization: There is no mention of sanitizing or escaping the data content before the agent processes or displays it.
  • [NO_CODE]: The skill does not include any scripts, binaries, or automated installation procedures. It is composed entirely of markdown documentation and usage examples for a pre-installed command-line utility.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 04:09 PM