PraisonAI MCP `tools/call` path-traversal and RCE via Python `.pth` injection
Key facts
- Attack vector
- Network
- Privileges required
- None
- User interaction
- Passive
- Attack complexity
- Low
- Attack requirements
- None
- Impact type
- Code execution
- Confidentiality · integrity · availability
- High · High · High
- Fix status
- Fixed
Attack profile from CVSS 4.0 · CNA. Fix status from CNA and GitHub. Impact type from CWE-829.
Summary
PraisonAI is a multi-agent teams system. Prior to version 4.6.34, PraisonAI's MCP (Model Context Protocol) server (praisonai mcp serve) registers four file-handling tools by default — praisonai.rules.create, praisonai.rules.show, praisonai.rules.delete, and praisonai.workflow.show. Each accepts a path or filename string from MCP tools/call arguments and joins it onto ~/.praison/rules/ (or, for workflow.show, accepts an absolute path) with no containment check. The JSON-RPC dispatcher passes params["arguments"] blind to each handler via **kwargs without validating against the advertised input schema. By setting rule_name="../../<some-path>" an attacker walks out of the rules directory and writes any file the running user can write. Dropping a Python .pth file into the user site-packages directory escalates this primitive to arbitrary code execution in any subsequent Python process the user spawns — the next praisonai CLI invocation, an IDE script run, the user's python REPL, or any background Python service. This issue has been patched in version 4.6.34.
Why it matters now
SSVC decision points
CISA ADP · SSVC 2.0.3
- Exploitation
- Poc
- Automatable
- No
- Technical impact
- Total
The decision (Track, Attend, Act) also depends on your mission and well-being impact; see CISA's SSVC guide.
Affected products
| Vendor | Product | Affected versions | Fixed in | Status | Source |
|---|---|---|---|---|---|
| MervinPraison | PraisonAI | before 4.6.34 | 4.6.34 | Patch available | CNA |
| — | PraisonAI (PyPI) | before 4.6.34 | 4.6.34 | Patch available | GHSA-9mqq-jqxf-grvw |
- PraisonAI VendorMervinPraison Affected before 4.6.34 Fixed in 4.6.34 Patch available CNA
- PraisonAI (PyPI) Vendor— Affected before 4.6.34 Fixed in 4.6.34 Patch available GHSA-9mqq-jqxf-grvw
CPE configurations NVD · 1
| CPE | Versions | Role | Vendor |
|---|---|---|---|
| cpe:2.3:a:praison:praisonai:*:*:*:*:*:*:*:* | before 4.6.34 | Vulnerable | MervinPraison |
- cpe:2.3:a:praison:praisonai:*:*:*:*:*:*:*:* Versions before 4.6.34 Vulnerable ·MervinPraison
Remediation
- Fix · CNA PraisonAI (MervinPraison) Fixed in 4.6.34
- Fix · GHSA-9mqq-jqxf-grvw PraisonAI (PyPI) Fixed in 4.6.34 Advisory →
Timeline
- 2026-10-02 Threatist record created from CVE-2026-44336
- 2026-10-03 Alias added: GHSA-9mqq-jqxf-grvw
- 2026-10-25 EPSS moved to the below the 50th percentile
References
More MervinPraison Praisonai vulnerabilities
- TI-2026-30083 Incomplete fix for CVE-2026-34935: Command Injection in MervinPraison/PraisonAI CVSS 3.1CRITICAL 9.8
- TI-2026-29070 PraisonAI: Critical RCE via `type: job` workflow YAML CVSS 3.1CRITICAL 9.8
- TI-2026-28940 PraisonAI has Server-Side Request Forgery via Unvalidated webhook_url in Jobs API CVSS 3.1CRITICAL 10.0
- TI-2026-28768 PraisonAI Affected by Remote Code Execution via YAML Deserialization in Agent Definition Loading CVSS 3.1CRITICAL 9.8
- TI-2026-27159 PraisonAI: OS Command Injection in MCPHandler.parse_mcp_command() CVSS 3.1CRITICAL 9.8
- TI-2026-27158 PraisonAI: Second-Order SQL Injection in `get_all_user_threads` CVSS 3.1CRITICAL 9.8
Sources
- CVE-2026-44336 · CVE Program
- CVE-2026-44336 · NVD
- FIRST EPSS
- GHSA-9mqq-jqxf-grvw · GitHub Advisory Database · CC BY 4.0
This product uses the NVD API but is not endorsed or certified by the NVD. Sources and licences