skill-optimize

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes local Python scripts (audit.py and apply.py) within the skill directory to perform usage analysis and settings modification.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and parses JSON data produced by an internal auditing script to determine which settings to modify, creating an injection surface.
  • Ingestion points: The stdout of python3 scripts/audit.py --json which is parsed by the agent.
  • Boundary markers: None explicitly present in the shell command construction; the agent is directed to parse the JSON object and pass extracted arrays to another script.
  • Capability inventory: The skill can execute shell commands and modify system settings via the apply.py script.
  • Sanitization: The workflow incorporates several "AskUserQuestion" steps for human verification before applying changes.
  • [DYNAMIC_EXECUTION]: The skill dynamically constructs shell commands by interpolating JSON-formatted data derived from the audit process into script arguments for apply.py.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 05:16 PM
Security Audit — agent-trust-hub — skill-optimize