A SQL injection vulnerability exists in MantisBT version 2.24.3 in the mc_project_get_users function, specifically via the 'access' parameter exposed through the SOAP API. Unsanitized input to this parameter is directly incorporated into SQL queries, allowing attackers to manipulate database queries.
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.
1 valid exploit after Mallory filtered fakes, detection scripts, and README-only repos (1 hidden).
This repository contains a single Python proof-of-concept exploit, CVE-2020-28413.py, targeting CVE-2020-28413 in MantisBT 2.24.3. The script is a real exploit rather than a detector: it crafts SOAP requests to the MantisConnect endpoint /api/soap/mantisconnect.php and injects SQL through the access parameter of the mc_project_get_users method. Repository structure is minimal: one standalone Python file with no framework integration. The code uses requests for HTTP POST delivery and lxml.etree to parse the XML SOAP response. Execution starts in main(), which first calls Cantidad_Usuarios_Mantis() to determine the number of users in mantis_user_table, then iterates through users, calling Obtener_Id() to recover each user ID and name, and brute_force() to extract each password hash one character at a time. Core exploit capability: authenticated SQL injection over SOAP. The script requires the operator to manually set the target base URL (home) and valid credentials (mantis_db_user and mantis_db_pass). It then sends crafted UNION and conditional SQL expressions embedded in the SOAP XML body. The extraction logic is tailored to MantisBT's mantis_user_table and assumes 32-character lowercase alphanumeric password hashes. The exploit does not deliver code execution or a shell; its purpose is credential/hash extraction from the backend database. Notable implementation details: SSL verification is disabled (verify=False), an intercepting proxy is preconfigured but commented out, and response parsing is brittle string-splitting over XML content rather than structured XPath parsing. Overall, this is an operational single-file exploit for database exfiltration against a specific vulnerable MantisBT SOAP interface.
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.
No public activity tracked yet. Mallory keeps watching.
No public activity observed for this vulnerability.
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.