PatchSiren cyber security CVE debrief
CVE-2026-56838 MervinPraison CVE debrief
CVE-2026-56838 is a high-severity vulnerability in PraisonAI, a multi-agent teams system, affecting versions from 4.5.87 to before 4.6.59. The issue arises from the recipe.run_stream() method bypassing dangerous-tool policy checks, potentially allowing the execution of denied tools. This vulnerability has been patched in version 4.6.59. The vulnerability impacts PraisonAI deployments, allowing execution of denied tools via the streaming path. Defenders should assess exposure and prioritize upgrading to version 4.6.59 or later.
- Vendor
- MervinPraison
- Product
- PraisonAI
- CVSS
- HIGH 7.8
- CISA KEV
- Not listed in stored evidence
- Original CVE published
- 2026-10-23
- Original CVE updated
- 2026-10-29
- Advisory published
- 2026-10-23
- Advisory updated
- 2026-10-29
Who should care
Defenders and administrators using PraisonAI versions between 4.5.87 and 4.6.59 should assess exposure and prioritize upgrading to version 4.6.59 or later. Systems using PraisonAI versions between 4.5.87 and 4.6.59 should be assessed for exposure. Defenders should review compensating controls for exposed systems while remediation is scheduled and verified.
Why it matters
CVE-2026-56838 is a high-severity vulnerability in PraisonAI that allows execution of denied tools via the recipe.run_stream() method. Defenders should prioritize upgrading to version 4.6.59 or later and assess exposure in systems using affected versions.
- Execution of denied tools via recipe.run_stream() method
- Potential bypass of security policies in PraisonAI
- Need for verification of affected systems and versions
- Priority on upgrading to patched version 4.6.59
Technical summary
The recipe.run_stream() method in PraisonAI versions 4.5.87 to before 4.6.59 bypasses dangerous-tool policy checks, potentially allowing execution of denied tools. This is patched in version 4.6.59. The vulnerability impacts PraisonAI deployments, allowing execution of denied tools via the streaming path. The issue arises from the recipe.run_stream() method loading the same recipe, checking dependencies, and then calling _execute_recipe() without running the dangerous-tool policy check. Defenders should prioritize upgrading to version 4.6.59 or later to mitigate this vulnerability.
Defensive priority
Defenders should prioritize upgrading to version 4.6.59 or later to mitigate this vulnerability. Systems using PraisonAI versions between 4.5.87 and 4.6.59 should be assessed for exposure.
Recommended defensive actions
- Upgrade to PraisonAI version 4.6.59 or later
- Assess PraisonAI versions between 4.5.87 and 4.6.59 for exposure
- Monitor systems using affected PraisonAI versions for potential misuse
- Review compensating controls for exposed systems while remediation is scheduled and verified
- Check relevant monitoring, detection, and logs for exposed assets that need extra review
- Track exceptions, retest remediated assets, and close the item only after evidence is documented
- Confirm whether affected product deployments exist in managed environments and assign an owner for follow-up
Evidence notes
The CVE record and source item provide details on the vulnerability, its impact, and the patched version. However, additional information on potential exploitation or affected systems is limited. There is no evidence of exploitation, but defenders should verify affected systems and versions. The source item and CVE record confirm the vulnerability details, but further information is needed to assess exposure.
Sources and references
Verified primary and authoritative sources
-
CVE-2026-56838 CVE Program record
Publisher, destination, and source semantics verified
URL: https://www.cve.org/CVERecord?id=CVE-2026-56838
CVE Program - Official CVE Program record with source-provided CVE metadata.
-
CVE-2026-56838 NVD vulnerability detail
Publisher, destination, and source semantics verified
URL: https://nvd.nist.gov/vuln/detail/CVE-2026-56838
NIST National Vulnerability Database - Official NIST NVD detail page and source-specific vulnerability assessment.
Supplemental references
-
PraisonAI recipe.run_stream skips dangerous-tool policy enforcement
Unverified legacy reference
URL: https://raw.githubusercontent.com/CVEProject/cvelistV5/main/cves/2026/56xxx/CVE-2026-56838.json
cve_program_cvelist_v5
-
Source reference
Unverified legacy reference
URL: https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-v847-hxxw-3pxg
Supplemental source - x_refsource_CONFIRM
Methodology and review provenance
AI-assisted synthesis based on stored public vulnerability evidence. System validation, approval state, and publication status do not by themselves establish human review of this revision. PatchSiren helps prioritize defensive review and does not prove exposure or remediation on any system.