Skills
skills/madappgang/mnemex/agentbench-eval/Gen Agent Trust Hub

agentbench-eval

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute various local Python scripts and shell scripts located within the repository infrastructure, such as run_condition.py, restore_indexes.sh, and evaluate.py, to manage benchmarking workflows and data restoration.
  • [SAFE]: The skill accesses sibling directories and uses well-known API providers (OpenRouter) for model access, which are appropriate and expected behaviors for a benchmark evaluation harness. No signs of data exfiltration, credential theft, or obfuscation were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 02:58 PM
Security Audit — agent-trust-hub — agentbench-eval