SSRF / data exfiltration in Splunk AI Toolkit default domain allowlist
CVE-2026-20265 affects Splunk AI Toolkit versions prior to 5.7.4. The issue is caused by an insecure default domain allowlist that does not properly restrict outbound AI agent HTTP requests to approved external domains. As a result, an attacker can cause the toolkit to initiate outbound HTTP requests to attacker-controlled infrastructure. The provided content indicates this can be triggered by a low-privileged user, though one mention context inconsistently states admin or power-role users; the primary description and advisory data indicate low-privileged non-admin, non-power users are sufficient. The vulnerability can expose data through outbound requests and is best characterized as a server-side request forgery style weakness arising from insufficient destination restriction.
Are you exposed to this one?
Mallory correlates every CVE against your assets, your vendors, and active adversary campaigns. Know which vulnerabilities matter for you, not just which ones are loud.
Impact, mitigation & remediation
What it means. What to do now. Patch path, mitigations, and the assume-compromise checklist.
Impact
What an attacker gets, and what they’ve been doing with it.
Mitigation
If you can’t patch tonight, do this now.
Remediation
Patch, then assume compromise.
Exploits
No public exploits tracked yet. Mallory keeps watching.
No public exploit code observed for this vulnerability.
Affected products & vendors
Products and vendors Mallory has correlated with this vulnerability. Open in Mallory to drill down to specific CPE configurations and version ranges.
Vendor-confirmed product mapping. Mallory continuously reconciles this list against your asset inventory.
Recent activity
4 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.
A medium-severity information disclosure vulnerability in Splunk AI Toolkit caused by an insecure default domain allowlist, enabling admin or power-role attackers to trigger outbound HTTP requests to attacker-controlled servers and exfiltrate data.
An insecure default domain allowlist vulnerability in the Splunk AI Toolkit that allows a low-privileged user to trigger outbound HTTP requests, potentially enabling data exfiltration to an attacker-controlled server.
The version that knows your environment.
Query your assets running an affected version, and investigate the blast radius.
Every observed campaign linking this CVE to a named adversary.
Malware families riding this exploit, with evidence and IOCs.
YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.
Cross-references every affected SKU, including bundled OEM variants.
Community discussion across Reddit, Mastodon, and other social sources.