CVE-2026-57143: PraisonAI: Server-Side Request Forgery (SSRF) in SearxNG / search_web tools via attacker-controlled searxng_url parameter

Published Oct 12, 2026
·
Updated

PraisonAI is a multi-agent teams system. Prior to praisonaiagents 1.6.62, searxngsearch and the searchweb SearxNG provider pass the LLM-controlled searxngurl parameter directly to requests.get without validating its scheme, host, port, or resolved address. Prompt-injected content can direct the default search tool to internal JSON services, distinguish open and closed internal ports, and attempt cloud-metadata requests, with returned JSON results entering agent context. This issue is fixed in praisonaiagents 1.6.62.

Affected Software

1 affected component
pypi/praisonaiagents<1.6.62

Remediation

Recommended actions to resolve this vulnerability, in priority order.

  1. Upgrade

    Upgrade praisonaiagents to a version that resolves this vulnerability.

    Fixed in 1.6.62

Event History

Oct 12, 2026
CVE Published
via MITRE·06:48 PM
Data Sourced
via MITRE·06:48 PM
DescriptionSeverityWeakness
Data Sourced
via NVD·07:16 PM
DescriptionSeverityWeakness

Frequently Asked Questions

1

Which deployments are exposed?

Deployments using praisonaiagents before 1.6.62 are exposed when they use the searxng_search tool or the search_web SearxNG provider. The affected searxng_url parameter is LLM-controlled, including through prompt-injected content.

2

What does an attacker need to exploit this issue?

An attacker needs a way to influence content processed by the LLM so it directs the search tool to an attacker-chosen searxng_url. No authentication or direct network access to the target service is required according to the supplied vector.

3

What can exploitation expose or enable?

The tool can send requests to internal JSON services, probe whether internal ports are open or closed, and attempt requests to cloud metadata endpoints. JSON returned by reachable services can be placed into the agent context.

4

What should teams do if they cannot patch immediately?

Restrict or disable use of the searxng_search tool and the SearxNG search_web provider where untrusted or prompt-injectable content can influence tool inputs. Limit the agent runtime's network access to internal services and cloud metadata endpoints.

5

What version contains the fix?

The issue is fixed in praisonaiagents 1.6.62. Versions prior to 1.6.62 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