Security

Security is a foundational principle in FixPanic. Since agents execute commands on your production infrastructure, we've built multiple layers of protection to keep your systems safe.

Security Philosophy

FixPanic follows the principle of defense in depth:

Command Whitelisting

Only pre-approved commands can be executed by agents.

Safe Commands (Auto-Approved)

These commands are generally safe and allowed by default:

Category	Examples
Process Info	`ps`, `top`, `pgrep`
System Info	`uname`, `hostname`, `uptime`
Disk Info	`df`, `du`, `lsblk`
Network Info	`netstat`, `ss`, `ip`
Log Reading	`cat`, `head`, `tail`, `grep` (on allowed paths)
Service Status	`systemctl status`
Container Info	`docker ps`, `docker logs`

Restricted Commands (Require Approval)

These commands require explicit human approval:

Category	Examples	Risk
Service Control	`systemctl restart`	Medium
Process Control	`kill -15`	Medium
Container Control	`docker stop`	Medium
File Modification	`touch`, `mkdir`	Medium

Blocked Commands (Never Allowed)

These commands are never executed, regardless of approval:

Category	Examples	Why Blocked
Destructive	`rm -rf`, `dd`	Data loss
Privilege Escalation	`sudo`, `su`, `chmod +s`	Security
Shell Escape	`bash -c`, `eval`, `cmd`	Bypass controls
Network Attack	`nmap`, `nc -l`	Malicious use
System Modification	`passwd`, `useradd`	Unauthorized access

Shell Operators

Shell operators are blocked to prevent command injection:

Operator	Blocked
`\|` (pipe)	Yes
`&&` (and)	Yes
`\|\|` (or)	Yes
`;` (sequence)	Yes
`>`, `>>` (redirect)	Yes
`<` (input redirect)	Yes
` (backtick)	Yes
`$()` (command sub)	Yes

Path Restrictions

Agents are blocked from accessing sensitive paths:

Blocked Paths

Path	Reason
`/etc/passwd`	User information
`/etc/shadow`	Password hashes
`/root`	Root home directory
`/home/*/.ssh`	SSH keys
`/var/lib/docker`	Docker data
`~/.aws`	AWS credentials
`~/.kube`	Kubernetes config

Allowed Paths

Path	Purpose
`/var/log`	System logs
`/tmp`	Temporary files
`/proc`	Process information
`/sys`	System information
Application directories	Your applications

Resource Limits

Commands are constrained to prevent resource abuse:

Limit	Default	Purpose
Execution Timeout	30 seconds	Prevent hanging
CPU Time	Limited	Prevent CPU abuse
Memory	Limited	Prevent memory exhaustion
Output Size	1 MB	Prevent output flooding

Note: These limits can be configured per-cluster for specific use cases. Contact support for custom limits.

Process Isolation

Commands run in isolated environments:

Sandboxing

Commands execute in separate processes
No shell expansion
Clean environment variables
Limited file descriptors
No network access from commands

Process Tree

Audit Logging

Every action is logged:

Logged Events

Event	Details Captured
Command Execution	Command, args, timestamp, user
Approval Decision	Approved/denied, who, when
Connection Events	Connect, disconnect, errors
Configuration Changes	What changed, who, when

Audit Log Format

{
  "timestamp": "2024-01-15T10:30:00Z",
  "event_type": "COMMAND_EXECUTED",
  "agent_id": "agent_abc123",
  "cluster_id": "cluster_xyz789",
  "user_id": "user_def456",
  "command": "ps aux",
  "result": "success",
  "execution_time_ms": 150
}