CVE-2026-57142: PraisonAI recipe workflow policy can be bypassed by declaring and YAML-approving dangerous tools outside TEMPLATE.yaml

Published Oct 23, 2026
·
Updated

PraisonAI is a multi-agent teams system. From version 4.5.87 to before version 4.6.61, PraisonAI recipe execution has a dangerous-tool policy that is supposed to block default-denied tools unless the caller explicitly passes allowdangeroustools=True. That policy only checks tools declared in TEMPLATE.yaml requires.tools. For steps-based recipes, the actual execution path loads workflow.yaml with YAMLWorkflowParser. That parser resolves agent-level tools: declarations and preserves top-level approve:. Workflow.start() then installs those YAML-approved tools into the approval context. As a result, an untrusted recipe can omit executecommand from TEMPLATE.yaml requires.tools, declare it in workflow.yaml agents..tools, and add top-level approve: [executecommand]. The caller did not set allowdangeroustools=True, but the recipe policy allows the recipe and the workflow approval path self-approves the critical shell tool. This issue has been patched in version 4.6.61.

Affected Software

1 affected component
PraisonAI PraisonAI>=4.5.87<4.6.61

Remediation

Recommended actions to resolve this vulnerability, in priority order.

  1. Upgrade

    Upgrade PraisonAI to a version that resolves this vulnerability.

    Fixed in 4.6.61

Event History

Oct 23, 2026
CVE Published
via MITRE·12:07 PM
Data Sourced
via MITRE·12:07 PM
DescriptionSeverityWeakness

Frequently Asked Questions

1

Who is exposed to this bypass?

Users who execute untrusted steps-based PraisonAI recipes are exposed if they use versions 4.5.87 through before 4.6.61. A malicious recipe can cause a dangerous shell-capable tool to be approved through its workflow.yaml configuration.

2

Does exploitation require the caller to enable dangerous tools?

No. The bypass works when the caller has not passed allow_dangerous_tools=True, because the recipe can omit the dangerous tool from TEMPLATE.yaml while declaring and approving it in workflow.yaml.

3

What must an attacker control to exploit this issue?

The attacker needs to supply or modify a steps-based recipe that is executed by PraisonAI. The recipe declares execute_command under workflow.yaml agent tools and adds it to the top-level approve list.

4

What version fixes the issue?

Version 4.6.61 includes the patch. Versions before 4.6.61, starting with 4.5.87, are affected.

Contact

SecAlerts Pty Ltd.
132 Wickham Terrace
Fortitude Valley,
QLD 4006, Australia
info@secalerts.co
By using SecAlerts services, you agree to our services end-user license agreement. This website is safeguarded by reCAPTCHA and governed by the Google Privacy Policy and Terms of Service. All names, logos, and brands of products are owned by their respective owners, and any usage of these names, logos, and brands for identification purposes only does not imply endorsement. If you possess any content that requires removal, please get in touch with us.
© 2026 SecAlerts Pty Ltd.
ABN: 70 645 966 203, ACN: 645 966 203