conversational-ads

Pass

Audited by Gen Agent Trust Hub on Oct 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues were identified during the analysis of this skill.
  • [COMMAND_EXECUTION]: The skill includes two Python scripts, answer_adjacent_copy_linter.py and conversational_ad_planner.py. Analysis of these scripts confirms they do not utilize subprocesses or system calls to execute external shell commands.
  • [DATA_EXFILTRATION]: No network-related functions (e.g., requests, urllib, socket) are present in the code. The scripts operate exclusively on local input files provided as CLI arguments.
  • [PROMPT_INJECTION]: A review of the instructions in SKILL.md and the reference documentation found no patterns intended to manipulate agent behavior, bypass safety filters, or extract system prompts.
  • [DYNAMIC_EXECUTION]: The scripts do not use dangerous functions such as eval(), exec(), or dynamic module loading. Data processing is restricted to standard JSON parsing.
  • [OBFUSCATION]: All files are written in clear, plain text. No Base64, hex-encoded strings, or hidden Unicode characters were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 27, 2026, 11:11 PM