8 matches found
CVE-2026-56838
PraisonAI is a multi-agent teams system. From version 4.5.87 to before version 4.6.59, PraisonAI recipe execution blocks default-denied dangerous tools unless the caller explicitly passes allowdangeroustools=True. The normal recipe.run path enforces this with checktoolpolicy. The streaming path,...
CVE-2026-56838 PraisonAI recipe.run_stream skips dangerous-tool policy enforcement
PraisonAI is a multi-agent teams system. From version 4.5.87 to before version 4.6.59, PraisonAI recipe execution blocks default-denied dangerous tools unless the caller explicitly passes allowdangeroustools=True. The normal recipe.run path enforces this with checktoolpolicy. The streaming path,...
CVE-2026-56838 PraisonAI recipe.run_stream skips dangerous-tool policy enforcement
PraisonAI is a multi-agent teams system. From version 4.5.87 to before version 4.6.59, PraisonAI recipe execution blocks default-denied dangerous tools unless the caller explicitly passes allowdangeroustools=True. The normal recipe.run path enforces this with checktoolpolicy. The streaming path,...
CVE-2026-56838 PraisonAI recipe.run_stream skips dangerous-tool policy enforcement
PraisonAI is a multi-agent teams system. From version 4.5.87 to before version 4.6.59, PraisonAI recipe execution blocks default-denied dangerous tools unless the caller explicitly passes allowdangeroustools=True. The normal recipe.run path enforces this with checktoolpolicy. The streaming path,...
CVE-2026-56838
PraisonAI versions 4.5.87 through 4.6.57 are vulnerable to a policy enforcement bypass in the recipe.run_stream() function. While the standard recipe.run() path correctly invokes _check_tool_policy() to block dangerous tools unless allow_dangerous_tools=True is specified, the streaming path calls...
EUVD-2026-108879
PraisonAI is a multi-agent teams system. From version 4.5.87 to before version 4.6.59, PraisonAI recipe execution blocks default-denied dangerous tools unless the caller explicitly passes allowdangeroustools=True. The normal recipe.run path enforces this with checktoolpolicy. The streaming path,...
Mervin Praison PraisonAI processing logic error vulnerability
A security vulnerability exists in Mervin Praison PraisonAI versions from 4.5.87 up to and including 4.6.59. This vulnerability arises because the recipe.runstream stream execution path fails to perform checks on dangerous tools, allowing recipes declared as executecommand to bypass restrictions ...
CVE-2026-56838
PraisonAI recipe execution blocks default-denied dangerous tools unless the caller explicitly passes "allowdangeroustools=True". The normal "recipe.run" path enforces this with "checktoolpolicy". The streaming path, "recipe.runstream", loads the same recipe, checks dependencies, and then calls...