Approvals & Whitelisting

FixPanic includes a powerful approval system that ensures you maintain control over what agents can do on your infrastructure.

How Approvals Work

When an agent proposes a potentially sensitive action, the system pauses execution and requests your approval before proceeding.

┌─────────────────────────────────────────────────────────────┐
│  Agent wants to execute:                                     │
│                                                              │
│  sudo systemctl restart nginx                                │
│                                                              │
│  This will restart the web server.                          │
│                                                              │
│  [Approve]  [Deny]  [Add to Whitelist]                      │
└─────────────────────────────────────────────────────────────┘

Approval Flow

1. Agent Proposes Action - You ask the agent to do something that requires a sensitive command.
2. System Pauses - The agent pauses and generates an approval request instead of executing.
3. Review Request - You see exactly what command or action the agent wants to perform.
4. Make Decision - You can Approve, Deny, or Whitelist the action.
5. Execution (if approved) - The agent proceeds with the action and reports the results.

Pending Approvals

Finding Pending Approvals

When there's a pending approval:

• A notification badge appears in the chat
• The conversation shows the pending request
• The agent mentions it's waiting for approval

Approval Details

Each approval request shows:

• Command: The exact command to be executed
• Context: Why the agent wants to run this
• Impact: Potential effects of the action
• Risk Level: Low, Medium, or High

Making Approval Decisions

Approve

Click "Approve" when:

• You understand the command
• You trust it's safe to execute
• You want the action to proceed

The agent will execute the command and continue.

Deny

Click "Deny" when:

• The command isn't what you intended
• You want to take a different approach
• You need more information first

The agent will acknowledge the denial and ask how to proceed.

Whitelist

Click "Add to Whitelist" when:

• You trust this specific command
• You want to approve it automatically in the future
• You're comfortable with it running without review

Warning: Whitelisted commands bypass approval for all future requests. Only whitelist commands you're certain are safe.

Command Whitelisting

What is Whitelisting?

Whitelisting allows specific commands to execute automatically without requiring approval.

How Whitelisting Works

Whitelisted commands:

• Execute immediately when requested
• Don't generate approval notifications
• Still appear in the activity log
• Can be removed from the whitelist at any time

Adding to Whitelist

You can whitelist commands in several ways:

During Approval:

• Click "Add to Whitelist" on an approval request

From Settings:

1. Go to Settings > Security > Whitelist
2. Click "Add Command"
3. Enter the command pattern
4. Click "Save"

Whitelist Patterns

You can whitelist:

• Exact commands: docker ps
• Command prefixes: docker *
• Commands with arguments: systemctl status *

Managing the Whitelist

View and manage your whitelist in Settings > Security > Whitelist:

Removing from Whitelist

1. Go to Settings > Security > Whitelist
2. Find the command you want to remove
3. Click the delete icon
4. Confirm removal

After removal, the command will require approval again.

Approval Presets

Presets allow you to apply pre-defined approval configurations quickly.

Available Presets

Applying a Preset

1. Go to Settings > Security > Presets
2. Select a preset
3. Click "Apply"
4. Confirm the change

Note: Applying a preset replaces your current whitelist. Make sure to review the preset's included commands before applying.

Risk Levels

Commands are categorized by risk:

Low Risk (Usually Auto-Approved)

Read-only commands that don't modify state:

• ps aux - List processes
• df -h - Check disk space
• cat /var/log/syslog - Read logs

Medium Risk (Default: Requires Approval)

Commands that could affect services:

• systemctl restart - Restart services
• docker stop - Stop containers
• kill -15 - Graceful process termination

High Risk (Always Requires Approval)

Dangerous or destructive commands:

• rm -rf - Delete files/directories
• systemctl disable - Disable services
• kill -9 - Force kill processes
• Database modifications

Hard-Blocked Commands

Some commands are never allowed, regardless of approval:

• Commands that could compromise security
• System-level destructive operations
• Commands accessing sensitive paths

Approval Best Practices

Do's

• Review carefully - Read the full command before approving
• Understand impact - Consider what the command will do
• Whitelist cautiously - Only whitelist commands you fully trust
• Use presets - Start with a preset and customize

Don'ts

• Don't rush - Take time to understand approval requests
• Don't over-whitelist - Keep the whitelist minimal
• Don't ignore denials - Understand why you denied something

Audit Trail

All approval decisions are logged:

• Approved commands with timestamps
• Denied requests with reasons
• Whitelist changes
• Who made each decision

Access the audit trail in Settings > Security > Audit Log.

Next Steps

• Advanced Features - Explore model selection, deep linking, and more.
• Security - Learn more about FixPanic's security model.