client-package-local

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs legitimate business operations: reading local markdown and CSV/JSON files, and creating new local directories and markdown/CSV files to organize deliverables.
  • [SAFE]: Google Sheets interactions are handled through the 'Rube' MCP server (using Composio Google Sheets tools). This is a standard integration for productivity tasks and does not involve direct credential handling within the skill.
  • [DATA_EXPOSURE]: The skill reads from local client folders (strategies, leads, content). While this involves potentially sensitive client data, it is the primary intended purpose of the skill (packaging client work). The data remains within the local filesystem or is uploaded to the user's own Google Sheets account.
  • [COMMAND_EXECUTION]: The skill uses mkdir -p to create local directories. This is a benign use of shell commands for file organization.
  • [SAFE]: No remote code execution, obfuscation, persistence mechanisms, or unauthorized privilege escalation patterns were detected. Dependencies are limited to the 'Rube' MCP server.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 06:25 PM
Security Audit — agent-trust-hub — client-package-local