CVE-2026-47419 is an insecure direct object reference vulnerability in PraisonAI Platform affecting versions prior to 0.1.4. The flaw exists in the agent CRUD endpoints for retrieving, updating, and deleting agents within workspaces. Authorization checks validate only that the requester is a member of the workspace identified in the URL, but the subsequent agent lookup is performed by primary key alone without constraining the lookup to the same workspace. As a result, an authenticated user with membership in one workspace can supply the identifier of an agent belonging to a different workspace and cause the application to operate on that foreign resource. This breaks tenant isolation and enables cross-workspace access to agent objects.
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.
What it means. What to do now. Patch path, mitigations, and the assume-compromise checklist.
What an attacker gets, and what they’ve been doing with it.
If you can’t patch tonight, do this now.
Patch, then assume compromise.
No public exploits tracked yet. Mallory keeps watching.
No public exploit code observed for this vulnerability.
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.
2 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.
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.