analyze

Pass

Audited by Gen Agent Trust Hub on Jul 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill uses dynamic context injection via the !date command to include a timestamp in the report. This is a benign use of shell execution for metadata purposes.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its core function involves processing untrusted external content (code and project files).
  • Ingestion points: Data is ingested through the $ARGUMENTS variable and file reading tools like grep and read during the reconnaissance phase.
  • Boundary markers: The instructions do not define clear boundaries or markers to distinguish between system instructions and analyzed content.
  • Capability inventory: The skill utilizes file system tools (glob, grep, read), task management (background_task), and document management (obsidian_append_content).
  • Sanitization: There are no explicit sanitization or validation steps for the data processed from the project files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 26, 2026, 12:10 AM
Security Audit — agent-trust-hub — analyze