alterlab-kvkk-dmp

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-supplied project metadata (e.g., project title, retention period) that is interpolated into a generated plan skeleton.
  • Ingestion points: User-provided arguments for the kvkk_dmp_scaffold.py script.
  • Boundary markers: The output is structured as a markdown file, but there are no specific markers to prevent instructions within user metadata from being interpreted by subsequent agents.
  • Capability inventory: The skill has Write and Bash access to run the localized Python script and save outputs.
  • Sanitization: The script performs standard string interpolation without advanced sanitization for markdown control characters. However, given the skill's purpose as a documentation generator, this risk is considered negligible.
  • [COMMAND_EXECUTION]: The skill utilizes a deterministic Python script (scripts/kvkk_dmp_scaffold.py) to generate plan skeletons. The script is self-contained, relies solely on the Python standard library, and performs no network or sensitive file operations.
  • [EXTERNAL_DOWNLOADS]: The skill references official Turkish government domains (e.g., mevzuat.gov.tr, kvkk.gov.tr, resmigazete.gov.tr) as authoritative sources for legal text. These references are appropriate and target trusted official institutions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 02:37 PM