CVE-2026-57137
CVE-2026-57137 · PraisonAI AgentLoop onToolCall approval runs after tool execution
Diese Seite enthält Fakten aus öffentlichen Sicherheitsquellen. Lesen Sie vor dem Aktualisieren die Originalmeldung.
PraisonAI is a multi-agent teams system. From 1.4.0 until 1.7.2, createAgentLoop() in src/praisonai-ts/src/ai/agent-loop.ts passes executable tools to generateText() before invoking the onToolCall approval callback. Because the wrapped AI SDK executes tool handlers during generation, a callback that returns false records tool_rejected only after the denied tool has already produced side effects and populated toolResults. Applications using onToolCall as a human or policy approval boundary can therefore execute rejected file, command, API, or data-modifying operations. This issue is fixed in version 1.7.2.
CVSS 8.8 · Hoch · EPSS 0,5% · Korrektur verfügbar
- Veröffentlicht
- Aktualisiert
Technologie-Stack: node