# CVE-2026-56836: PraisonAI authentication bypass in Recipe Serve command

Source: https://junglewise.ai/threats/cve-2026-56836-praisonai-recipe-serve-typer-command-bypasses-the-non-localhost  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-10-23. Updated: 2026-10-23  
License: CC BY 4.0 with a link back

CVE-2026-56836 · Severity: high · CVSS 8.2 · Published 2026-10-23

PraisonAI is a multi-agent teams system for building AI applications. The Recipe Serve command can expose an HTTP server on public network interfaces without requiring authentication, allowing unauthenticated attackers to access the Recipe API and administrative functions like /admin/reload. An attacker on the network can reconfigure or manipulate the running agent system without any credentials.

Technologies: [MervinPraison PraisonAI](https://junglewise.ai/threats/technologies/mervinpraison-praisonai), [praisonai (PyPI)](https://junglewise.ai/threats/technologies/pypi-praisonai). Vendors: [MervinPraison](https://junglewise.ai/threats/vendors/mervinpraison), [PyPI](https://junglewise.ai/threats/vendors/pypi).

## Quick answers

**Is CVE-2026-56836 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-56836 being exploited in the wild.

**How severe is CVE-2026-56836?** High severity, with a CVSS score of 8.2, published 23 October 2026.

**Which products and versions are affected?** MervinPraison PraisonAI 4.5.112 to 4.6.58; praisonai (PyPI) ECOSYSTEM:introduced=4.5.112,fixed=4.6.59.

**Is there a fix for CVE-2026-56836?** Yes. praisonai (PyPI): upgrade to 4.6.59 or later. Until then: Ensure the --api-key argument is explicitly provided when using --host with a non-localhost address (e.g., 0.0.0.0), or restrict binding to 127.0.0.1. Update to PraisonAI 4.6.59 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-56836 attacked?** It is attacked over the network.

**Am I affected by CVE-2026-56836?** Check whether you run MervinPraison PraisonAI or praisonai (PyPI) at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## How to fix

- praisonai (PyPI): upgrade to 4.6.59 or later

  ```sh
  pip install "praisonai>=4.6.59"
  ```


Ensure the --api-key argument is explicitly provided when using --host with a non-localhost address (e.g., 0.0.0.0), or restrict binding to 127.0.0.1. Update to PraisonAI 4.6.59 or later.

## Technical details

The vulnerable Typer command in src/praisonai/cli/commands/recipe.py accepts a --host parameter (defaulting to 127.0.0.1) and only enables authentication middleware when --api-key is explicitly passed. When --host is set to 0.0.0.0 or another non-localhost address without --api-key, the resulting Flask/Starlette app launches without any auth middleware, bypassing the hardened non-localhost guard present in the legacy recipe handler. Unauthenticated network requests can then reach the Recipe API and admin endpoints.

## Affected products

- MervinPraison PraisonAI 4.5.112 to 4.6.58

## Identifiers

- CVE-2026-56836
- GHSA-5qw8-f2g9-ff29
- PYSEC-2026-3502

## Timeline

- 2026-06-17: disclosed
- 2026-10-23: advisory
- 2026-06-17: patched: Version 4.6.59 includes the fix

## References

- [https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-5qw8-f2g9-ff29](https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-5qw8-f2g9-ff29)

## Related threats

- [PraisonAI authentication bypass in A2U event streaming](https://junglewise.ai/threats/cve-2026-57146-praisonai-a2u-incomplete-authentication-fix-leaves-current-serve)
- [PraisonAI sandbox escape when Landlock unavailable](https://junglewise.ai/threats/cve-2026-57144-praisonai-sandlocksandbox-falls-back-to-unrestricted-subprocess)
- [PraisonAI recipe policy bypass via YAML-approved tools](https://junglewise.ai/threats/cve-2026-57142-praisonai-recipe-workflow-policy-can-be-bypassed-by-declaring-and)
- [PraisonAI shell command injection in compute-bridged file tools](https://junglewise.ai/threats/cve-2026-57117-praisonai-compute-bridged-file-tools-allow-shell-command)
- [PraisonAI AgentOS authentication bypass in FastAPI deployment](https://junglewise.ai/threats/cve-2026-57116-praisonai-agentos-remains-unauthenticated-after-incomplete-fix)