Skills
Navigate
skills/fatih-developer/fth-skills/ecosystem-reasoning/Gen Agent Trust Hub

ecosystem-reasoning

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a management script scripts/install_all.py which executes shell commands via subprocess.run. This script is used to interact with the skills CLI tool to manage ecosystem components.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of member skills from the author's GitHub repository (fatih-developer/fth-skills) using npx. This is part of the intended lifecycle management for the ecosystem.
  • [COMMAND_EXECUTION]: Instructions in SKILL.md direct the agent to perform pre-installation checks and explicitly require user consent before executing the installation command (--execute flag).
  • [INDIRECT_PROMPT_INJECTION]: The skill routes user requests to different workflows. It implements a validation step in scripts/install_all.py that checks workflow IDs against a static configuration file (references/workflows.json), effectively preventing arbitrary input from reaching the shell execution layer.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 10:52 PM