CVE-2026-25996
📋 TL;DR
This vulnerability in Inspektor Gadget allows malicious containers to inject ANSI escape sequences into terminal output, potentially enabling terminal manipulation attacks. It affects operators using the default columns output mode interactively. The risk is limited to terminal effects rather than code execution.
💻 Affected Systems
- Inspektor Gadget
📦 What is this software?
Inspektor Gadget by Linuxfoundation
⚠️ Risk & Real-World Impact
Worst Case
Terminal takeover allowing command injection, screen manipulation, or data exfiltration through crafted escape sequences
Likely Case
Terminal corruption, screen clearing, cursor repositioning, or visual disruption of output
If Mitigated
Minor visual artifacts or no impact if output sanitization is enabled
🎯 Exploit Status
Requires ability to create malicious containers and target operators using ig interactively
🛠️ Fix & Mitigation
✅ Official Fix
Patch Version: v0.49.1
Vendor Advisory: https://github.com/inspektor-gadget/inspektor-gadget/security/advisories/GHSA-34r5-6j7w-235f
Restart Required: No
Instructions:
1. Update Inspektor Gadget to v0.49.1 or later
2. No restart required - fix applies to new ig run sessions
🔧 Temporary Workarounds
Use JSON output mode
allSwitch to JSON output which sanitizes control characters
ig run --output json
Disable interactive columns mode
allForce non-interactive mode or use different output format
ig run --output custom-columns
🧯 If You Can't Patch
- Restrict container creation to trusted users only
- Use ig only in non-interactive modes or with trusted workloads
🔍 How to Verify
Check if Vulnerable:
Check if running Inspektor Gadget version <0.49.1 and using ig run interactively
Check Version:
ig version
Verify Fix Applied:
Confirm version is ≥0.49.1 and test that escape sequences in string fields are properly sanitized
📡 Detection & Monitoring
Log Indicators:
- Unusual terminal behavior during ig operations
- ANSI escape sequences in ig output logs
Network Indicators:
- Not network exploitable - local terminal attack
SIEM Query:
Search for ig run commands with suspicious container activity patterns