CVE-2026-57137 PraisonAI AgentLoop onToolCall approval runs after tool execution
Attackers can bypass tool approval checks in PraisonAI because tools execute before the approval callback runs, allowing unauthorized operations.
- Exploited in the wild
- Not known
- Login needed
- Any signed-in user
- Attackable
- Over the internet
- Victim action needed
- Not needed
- Fix
- Available
What is CVE-2026-57137?
CVE-2026-57137 is a high-severity missing authorization vulnerability in MervinPraison PraisonAI, published on 15 Sep 2026. Any signed-in user can exploit it over the network, with no action from the victim.
Official description
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.
Which versions are affected by CVE-2026-57137?
2 open source packages ship an affected version, including praisonai.
| Package | Affected | Fixed in |
|---|---|---|
praisonai npm | From 1.4.0 | 1.7.2 |
praisonai npm | From 1.4.0 | 1.7.2 |
How do I fix CVE-2026-57137?
Upgrade praisonai to 1.7.2 or later.
Upgrade, and check permissions on every request on the server, not only in the interface. More about Missing authorization.
Is CVE-2026-57137 being exploited?
It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-57137 exists.
Sources
Also known as GHSA-h2w2-v7j6-xqm4.