Skip to content
KitploitKITPLOIT
ToolsExploitsBlog
Log in
Submit
ToolsExploitsBlog
Submit

Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!

Kitploit is a directory of hacking, cybersecurity, and pentesting tools. Discover the latest project updates to find vulnerabilities, analyze systems, automate testing, and strengthen your security.

FeedsContactPrivacy© 2026 Kitploit

Tool Directory

Categories

View all categories
Loading categories
CVE-2026-98432-Wazuh- — Wazuh SCA policy that audits host-level interim mitigations for CVE-2026-98432, checking nftables/iptables IP-option drop rules and TCP migration and MPTCP sysctls. | Kitploit
Tools/GitHubGitHub/infokom-ki/cve-2026-98432-wazuh-
Defensive ToolsVulnerability AnalysisConfiguration AuditingNetwork SecurityMisconfigurationIncident Response
GitHubinfokom-ki/cve-2026-98432-wazuh-

CVE-2026-98432-Wazuh-

Share

Wazuh SCA policy that audits host-level interim mitigations for CVE-2026-98432, checking nftables/iptables IP-option drop rules and TCP migration and MPTCP sysctls.

View Repository
4115 days agoNot yet reviewed

CVE-2026-98432 interim mitigation SCA policy

This policy checks host-level interim controls for CVE-2026-98432. It does not detect the vulnerability and does not prove that the running kernel is patched. Pair it with Wazuh Vulnerability Detection and a reboot check.

Checks

IDCheckPasses whenLimits
990100nftables drops inbound IPv4 packets carrying optionsnft list ruleset has a line with ip hdrlength != 5 or ip hdrlength > 5 and dropRule text only. The rule may sit in the wrong chain, after an accept rule, or not persist across reboot. Not applicable when nft is unavailable.
990101iptables drops inbound IPv4 packets carrying optionsiptables-save has a line with u32, the IHL match, and -j DROPRule text only. The rule may sit in the wrong chain, after an accept rule, or not persist across reboot. Not applicable when iptables-save is unavailable.
990102TCP request socket migration is disabled/proc/sys/net/ipv4/tcp_migrate_req is 0Runtime value only. Does not verify persistence or cover a BPF reuseport migration program. Not applicable when the sysctl is absent.
990103MPTCP is disabled/proc/sys/net/mptcp/enabled is 0Runtime value only. Does not verify persistence. Removes one exploitation route, not the use-after-free. MPTCP may be required by applications. Not applicable when MPTCP is absent.

nft list ruleset prints gt as >. Check 990100 matches both forms. iptables-save normalizes the u32 expression to hex. Check 990101 matches the normalized form and the decimal form.

If MPTCP is required, remove check 990103. The remaining checks still cover the IP option drop and the migration path.

A pass means the named control was present at scan time. It does not guarantee that the control persists or that the kernel is fixed. A fail means the control was absent or did not match. Not applicable means SCA could not evaluate the target; it is not a pass.

The policy includes a Linux requirement, so it does not run on non-Linux agents. Wazuh excludes not-applicable checks from the policy score. Read the individual results. A host with no firewall tooling can show a high score while both firewall checks are not applicable.

SCA reads command output in chunks. On hosts with very large nftables or iptables rulesets, the line-based firewall match can be unreliable near the chunk boundary. Verify those hosts manually.

What the policy cannot assess

  1. Kernel patch state: whether the installed kernel package is affected, whether the running kernel contains the fix, whether a livepatch is active, and whether a reboot is pending.
  2. Network exposure: whether an IPv4 TCP listener is reachable from an untrusted network, and whether a cloud security group, network ACL, or upstream firewall drops the packets.
  3. Firewall effectiveness: whether the rule is reached before an accept rule, attached to an input hook, or covers every interface.
  4. Persistence: whether the firewall rule or sysctl survives a reboot.
  5. Detection: whether a BPF reuseport migration program is active, whether an application disables MPTCP through GODEBUG=multipathtcp=0, whether exploitation was attempted or succeeded, and whether a host is immune.

If a command runs but exits non-zero, the SCA engine evaluates its output. That case is reported as failed when no pattern matches, not as not applicable. Check the agent log if a firewall check fails unexpectedly.

Deployment

Create a temporary agent group from hosts that Vulnerability Detection or the vendor tracker flagged as running an affected kernel. Do not add this policy to a permanent baseline.

  1. Place the file on the manager:
install -m 0640 cve_2026_98432_mitigation.yml /var/ossec/etc/shared/<group>/
chown wazuh:wazuh /var/ossec/etc/shared/<group>/cve_2026_98432_mitigation.yml
  1. Add the policy to /var/ossec/etc/shared/<group>/agent.conf:
<agent_config>
  <sca>
    <policies>
      <policy>etc/shared/cve_2026_98432_mitigation.yml</policy>
    </policies>
  </sca>
</agent_config>
  1. If the manager pushes the policy and it contains command checks, set this on each agent:
echo "sca.remote_commands=1" >> /var/ossec/etc/local_internal_options.conf

That setting lets the manager run commands on the endpoint. Restrict manager access before enabling it.

  1. Restart the agent:
systemctl restart wazuh-agent
  1. Confirm the scan result in the Configuration Assessment module and check /var/ossec/logs/ossec.log for policy errors. The command checks need nft or iptables-save in the agent PATH and permission to read the ruleset; the Linux agent normally runs as root. A containerized agent may see a different network namespace.

Check IDs 990100 to 990103 must be unique across all policy files. Change them if they collide with an existing policy.

Validation

python3 validate_policy.py

The validator requires PyYAML. It checks YAML structure, policy fields, requirements, rule syntax, unique check IDs, compliance and reference shape, and 32 fixture cases plus three requirements assertions.

The fixtures cover:

  • nftables != 5, real > 5 output, and the older gt 5 form
  • an nftables accept rule followed by a separate drop rule, which must fail
  • a comment line that contains the nftables pattern, which must fail
  • iptables u32 rules in normalized hex, leading-zero hex, and decimal forms, including a lower-case verdict
  • an iptables accept rule with a DROP comment, which must fail
  • an iptables drop rule without the u32 matcher, which must fail
  • a comment line that contains the iptables pattern, which must fail
  • missing commands, execution failures, and non-zero exits with empty output
  • tcp_migrate_req values 0, 1, and absent
  • mptcp.enabled values 0, 1, and absent
  • one firewall tool present, the other absent
  • fully mitigated, fully unmitigated, and no-firewall-tool environments
  • the Linux requirements check

This is not a Wazuh runtime test. The fixtures use Python re, while Wazuh uses OSRegex by default. The policy patterns avoid advanced regex constructs, so expected results should match, but run one lab agent before rolling out.

A kernel release string is not a reliable patch indicator. Distribution backports, ABI suffixes, livepatches, and a pending reboot all break version matching. See SCA-ASSESSMENT.md for the reasoning and the alternatives.

Removal

After the fleet runs a fixed kernel, remove the policy from agent.conf, delete the file from the group directory, and restart the agents. A mitigation policy that outlives the patch produces false failures and score noise.

Download Tool

Most Popular

View all →

Discover the most used tools by our community.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 The ultimate OSINT toolkit: investigation graph, confidence scoring, alias families, map, timeline and reports. 49 modules, 400+ sites, no API keys required.

    962
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    GhostLock One-Tap Execution App (CVE-2026-43499)

    558
  • 3

    redforge

    cybersecurity91/redforge

    Browser-side red team toolkit with 102 offline tools for recon, payload generation, credential attacks, crypto, encoding and engagement reporting.

    348
  • 4

    looking-glass

    anonghosty/looking-glass

    CSIRT platform that ingests Shadowserver reports into MongoDB and replays them as interactive maps, analysis, findings, advisories and executive dashboards.

    344
  • 5

    ARTEX

    baarai/artex

    LLM multi-agent autonomous penetration testing platform with planner/worker agents, asset and exploration graphs, MITM traffic recording, and vulnerability retesting.

    324
  • 6

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next.

    297

Explore all tools

Browse our collection of tools

View all tools →