CVE-2026-55522 is a code injection vulnerability (CWE-94) in MervinPraison's PraisonAI multi-agent teams system. The workflow "include" feature's _execute_include() method unsafely imports and executes tools.py from included recipes using importlib.util.spec_from_file_location() and spec.loader.exec_module() without honoring opt-in gates (PRAISONAI_ALLOW_TEMPLATE_TOOLS/PRAISONAI_ALLOW_LOCAL_TOOLS) or routing through the centralized safe loader. This flaw allows arbitrary module-level Python code execution during include setup if an attacker controls the included local recipe directory. The vulnerability bypasses previous hardening measures applied to automatic tools.py loading. It affects praisonai versions 3.9.26 through 4.6.57 and praisonaiagents versions 0.12.12 through 1.6.57. The vulnerability is fixed in praisonai 4.6.58 and praisonaiagents 1.6.58.

An attacker able to cause a victim process to run a workflow or recipe that includes an attacker-controlled local recipe can achieve arbitrary Python code execution with the privileges of the PraisonAI process user. This can lead to full compromise of the affected system. The CVSS 3.1 base score is 7.8 (high), reflecting local attack vector with low complexity, no privileges required, user interaction required, and high impact on confidentiality, integrity, and availability.

A fix is available in PraisonAI version 4.6.58 and praisonaiagents version 1.6.58. Users should upgrade to these versions or later to remediate this vulnerability. No other mitigations are indicated in the provided data.