$ loading_
Securely handle files and simulate commands with policy checks and approvals
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "permission-aware-mcp" yet — see the docs or source repo.
Scan log files in /var/log/app under read-only permissions, extract ERROR entries from the last 24 hours, and summarize them into a Markdown report sorted by time. If any access exceeds policy, pause and request approval first.
A time-sorted error log summary report, with an approval prompt if needed.
Review the timeout setting in config/settings.json, change timeout from 30 to 60, and show the diff, risk assessment, and required permissions before writing. Only perform the write after human approval.
A change preview, risk explanation, and approval workflow, followed by a safe write after approval.
Simulate commands to clean temporary files in ./tmp and ./cache. First provide the commands that would run, the files that may be removed, the risk level, and whether human confirmation is required. Do not actually delete anything.
A command simulation report with deletion candidates, risk assessment, and confirmation guidance.
Safely let AI read, write, move, copy, and manage files.
Restrict AI file access to one folder with safe read-only defaults.
Secure file and directory operations for autonomous AI development workflows.
Govern MCP tool calls by blocking dangerous and destructive actions before execution.
Execute shell commands securely over MCP for controlled dev and ops workflows.
Securely run AppleScript for AI-driven macOS app and system automation.