CVE-2026-56838
Deferred
Deferred - Pending Action
PraisonAI Recipe Execution Allows Dangerous Tools via Streaming Path
Vulnerability report for CVE-2026-56838, including description, CVSS score, EPSS score, affected products, exploitability, helpful resources, and attack-flow context.
Publication date: 2026-10-23
Last updated on: 2026-10-23
Assigner: GitHub, Inc.
Description
Description
PraisonAI is a multi-agent teams system. From version 4.5.87 to before version 4.6.59, PraisonAI recipe execution blocks default-denied dangerous tools unless the caller explicitly passes allow_dangerous_tools=True. The normal recipe.run() path enforces this with _check_tool_policy(). The streaming path, recipe.run_stream(), loads the same recipe, checks dependencies, and then calls _execute_recipe() without running the dangerous-tool policy check. As a result, a recipe that honestly declares execute_command in TEMPLATE.yaml requires.tools is denied by recipe.run(), but reaches the execution engine through recipe.run_stream() with allow_dangerous_tools=False. This issue has been patched in version 4.6.59.
CVSS Scores
EPSS Scores
| Probability: | |
| Percentile: |
Meta Information
Affected Vendors & Products
| Vendor | Product | Version / Range |
|---|---|---|
| MervinPraison | PraisonAI | >= 4.5.87, < 4.6.59 |
Helpful Resources
Exploitability
CWE
KEV
| CWE ID | Description |
|---|---|
| CWE-863 | The product performs an authorization check when an actor attempts to access a resource or perform an action, but it does not correctly perform the check. |
| CWE-78 | The product constructs all or part of an OS command using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the intended OS command when it is sent to a downstream component. |
| CWE-693 | The product does not use or incorrectly uses a protection mechanism that provides sufficient defense against directed attacks against the product. |
