Security Digest · Vulnerability record
CVE-2026-34938: PraisonAI: Python Sandbox Escape via str Subclass startswith() Override in execute_code
PraisonAI is a multi-agent teams system. Prior to version 1.5.90, executecode() in praisonai-agents runs attacker-controlled Python inside a three-layer sandbox that can be fully bypassed by passing a str subclass with an overridden startswith() method to the safegetattr wrapper, achieving arbitrary OS command…
CVSS 10.0 Critical EPSS 0.8% · p57
Should you act?
Critical and remotely reachable
No known exploitation yet, but it is network-reachable without credentials. This signal is computed from CISA KEV, EPSS and CVSS data; it is not an editorial verdict.
- Not in CISA KEV, but CISA’s SSVC assessment records a public proof-of-concept, so attack code exists.
- EPSS 0.8%, percentile 57: FIRST’s model estimates a 0.8% chance of exploitation activity in the next 30 days, higher than 57% of all scored CVEs (moderate). EPSS is a probability, not a verdict.
- Remotely reachable with no login and no user interaction. If the vulnerable service is exposed to the internet, prioritize it.
- Only applies if your dependency tree includes
praisonaiagents(PyPI) <= 1.5.89, fixed in 1.5.90. Check your lockfile, including transitive dependencies.
What it is
PraisonAI is a multi-agent teams system. Prior to version 1.5.90, executecode() in praisonai-agents runs attacker-controlled Python inside a three-layer sandbox that can be fully bypassed by passing a str subclass with an overridden startswith() method to the safegetattr wrapper, achieving arbitrary OS command execution on the host. This issue has been patched in version 1.5.90.
From the GitHub advisory
Summary
executecode() in praisonai-agents runs attacker-controlled Python inside a three-layer sandbox that can be fully bypassed by passing a str subclass with an overridden startswith() method to the safegetattr wrapper, achieving arbitrary OS command execution on the host.
Details
pythontools.py:20 (source) -> pythontools.py:22 (guard bypass) -> pythontools.py:161 (sink)
PoC
Weakness type: CWE-693.
Product: praisonaiagents (PyPI) · Assigned by: GitHub_M · Published: 2026-04-03
Affected and fixed versions
| Ecosystem | Package | Vulnerable | First fixed |
|---|---|---|---|
| PyPI | praisonaiagents | <= 1.5.89 | 1.5.90 |
| Vendor | Product | Affected versions (per CVE record) |
|---|---|---|
| MervinPraison | PraisonAI | < 1.5.90 |
Severity scores
| Scored by | CVSS | Score | Vector |
|---|---|---|---|
| GitHub_M | 3.1 | 10 Critical | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H |
| GitHub | 3.1 | 10.0 | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H |
- Attack vector: network (remotely reachable)
- Attack complexity: low
- Privileges required: none (no login)
- User interaction: none
- Scope: changed (can affect other components)
CVSS measures how bad a flaw is if exploited, not how likely exploitation is. Read it next to the exploitation data below.
Exploitation and urgency
- CISA KEV: not listed (no confirmed in-the-wild exploitation known to CISA).
- CISA SSVC assessment: exploitation poc, automatable yes, technical impact total.
- FIRST EPSS (model of 2026-10-21): 0.84% probability of exploitation activity in the next 30 days; percentile 57.
Timeline
- — GitHub-reviewed advisory GHSA-6vh2-h83c-9294 published (critical)
- — CVE record published by GitHub_M
- — CVE record last updated
Sources and references
Original sources first. Version and fix details should always be confirmed against the vendor or project advisory.
Vulnerability records
- GitHub Advisory GHSA-6vh2-h83c-9294 github.com
- CVE.org record CVE-2026-34938 cve.org
- NVD entry CVE-2026-34938 nvd.nist.gov
Other references from the CVE record and advisory
- github.com github.com