mcp-toggle-normalize
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill uses Python scripts executed via shell heredocs (`python3
- <<'EOF'`) to implement the logic for updating the JSON configuration. This is a common pattern for complex file manipulation within agent skills.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill accesses
~/.claude.jsonto manage project-specific connector settings. While this file contains environment configuration, the skill's operations are strictly local, and no network exfiltration patterns or hardcoded credentials were identified. - [SAFE]: The skill follows safety best practices by requiring a manual backup of the configuration file (
cp ~/.claude.json ~/.claude.json.bak-mcp-normalize) before applying changes and provides verification steps to ensure the file remains valid and compliant.
Audit Metadata