Romain Deperne

9 exploits Active since Mar 2026
CVE-2026-40864 GITHUB MEDIUM html WORKING POC
JupyterHub: Cross-origin form POSTs bypass XSRF
JupyterHub is software that allows users to create a multi-user server for Jupyter notebooks. In versions 4.1.0 through 5.4.4, XSRF protection (updated in 4.1.0) inappropriately treated requests with Sec-Fetch-Mode: no-cors as same-origin requests, bypassing XSRF checks. The JSON API is not affected, only HTTP form endpoints, such as /hub/spawn and /hub/accept-share, meaning attackers could trigger server spawn (but not access the server) and if the attacker is a JupyterHub user permitted to share access to their server, cause a user to accept a share and have access to the attacker's server. This issue has been fixed in version 5.4.5. If developers are unable to immediately upgrade, they can temporarily mitigate this issue by dropping requests to JupyterHub with Sec-Fetch-Mode: no-cors if they are using a reverse proxy.
CVSS 5.4
CVE-2026-41490 GITHUB HIGH python WORKING POC
Dagster Vulnerable to SQL Injection via Dynamic Partition Keys in Database I/O Manager Integrations
Dagster is an orchestration platform for the development, production, and observation of data assets. Prior to Dagster Core version 1.13.1 and prior to Dagster libraries version 0.29.1, the DuckDB, Snowflake, BigQuery, and DeltaLake I/O managers constructed SQL WHERE clauses by interpolating dynamic partition key values into queries without escaping. A user with the Add Dynamic Partitions permission could create a partition key that injects arbitrary SQL, which would execute against the target database backend under the I/O manager's credentials. Only deployments that use dynamic partitions are affected. Pipelines using static or time-window partitions are not impacted. This issue has been patched in Dagster Core version 1.13.1 and Dagster libraries version 0.29.1.
CVSS 8.3
CVE-2026-34940 GITHUB HIGH WORKING POC
KubeAI <0.23.2 Ollama Model URL - OS Command Injection
KubeAI is an AI inference operator for kubernetes. Prior to 0.23.2, the ollamaStartupProbeScript() function in internal/modelcontroller/engine_ollama.go constructs a shell command string using fmt.Sprintf with unsanitized model URL components (ref, modelParam). This shell command is executed via bash -c as a Kubernetes startup probe. An attacker who can create or update Model custom resources can inject arbitrary shell commands that execute inside model server pods. This vulnerability is fixed in 0.23.2.
CVSS 8.7
CVE-2026-34160 GITHUB HIGH WRITEUP
Chamilo LMS: Unauthenticated SSRF via PENS Plugin allows attacker to probe internal network and reach cloud metadata services
Chamilo LMS is an open-source learning management system. In versions prior to 2.0.0-RC.3, the PENS (Package Exchange Notification Services) plugin endpoint at public/plugin/Pens/pens.php is accessible without authentication and accepts a user-controlled package-url parameter that the server fetches using curl without filtering private or internal IP addresses, enabling unauthenticated Server-Side Request Forgery (SSRF). An attacker can exploit this to probe internal network services, access cloud metadata endpoints (such as 169.254.169.254) to steal IAM credentials and sensitive instance metadata, or trigger state-changing operations on internal services via the receipt and alerts callback parameters. No authentication is required to exploit either SSRF vector, significantly increasing the attack surface. This issue has been fixed in version 2.0.0-RC.3.
CVSS 8.6
CVE-2026-27825 GITHUB CRITICAL python WORKING POC
MCP Atlassian <0.17.0 - Path Traversal
MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Prior to version 0.17.0, the `confluence_download_attachment` MCP tool accepts a `download_path` parameter that is written to without any directory boundary enforcement. An attacker who can call this tool and supply or access a Confluence attachment with malicious content can write arbitrary content to any path the server process has write access to. Because the attacker controls both the write destination and the written content (via an uploaded Confluence attachment), this constitutes for arbitrary code execution (for example, writing a valid cron entry to `/etc/cron.d/` achieves code execution within one scheduler cycle with no server restart required). Version 0.17.0 fixes the issue.
CVSS 9.0
CVE-2026-33715 GITHUB HIGH python WORKING POC
Chamilo LMS has Unauthenticated SSRF and Open Email Relay via install.ajax.php test_mailer action
Chamilo LMS is an open-source learning management system. In version 2.0-RC.2, the file public/main/inc/ajax/install.ajax.php is accessible without authentication on fully installed instances because, unlike other AJAX endpoints, it does not include the global.inc.php file that performs authentication and installation-completed checks. Its test_mailer action accepts an arbitrary Symfony Mailer DSN string from POST data and uses it to connect to an attacker-specified SMTP server, enabling Server-Side Request Forgery (SSRF) into internal networks via the SMTP protocol. An unauthenticated attacker can also abuse this to weaponize the Chamilo server as an open email relay for phishing and spam campaigns, with emails appearing to originate from the server's IP address. Additionally, error responses from failed SMTP connections may disclose information about internal network topology and running services. This issue has been fixed in version 2.0.0-RC.3.
CVSS 7.2
CVE-2026-32247 GITHUB HIGH python WORKING POC
graphiti-core < 0.28.2 - Cypher Injection via SearchFilters.node_labels
Graphiti is a framework for building and querying temporal context graphs for AI agents. Graphiti versions before 0.28.2 contained a Cypher injection vulnerability in shared search-filter construction for non-Kuzu backends. Attacker-controlled label values supplied through SearchFilters.node_labels were concatenated directly into Cypher label expressions without validation. In MCP deployments, this was exploitable not only through direct untrusted access to the Graphiti MCP server, but also through prompt injection against an LLM client that could be induced to call search_nodes with attacker-controlled entity_types values. The MCP server mapped entity_types to SearchFilters.node_labels, which then reached the vulnerable Cypher construction path. Affected backends included Neo4j, FalkorDB, and Neptune. Kuzu was not affected by the label-injection issue because it used parameterized label handling rather than string-interpolated Cypher labels. This issue was mitigated in 0.28.2.
CVSS 8.1
CVE-2026-33980 GITHUB HIGH python WORKING POC
Azure Data Explorer MCP Server <=0.1.1 - KQL Injection
Azure Data Explorer MCP Server is a Model Context Protocol (MCP) server that enables AI assistants to execute KQL queries and explore Azure Data Explorer (ADX/Kusto) databases through standardized interfaces. Versions up to and including 0.1.1 contain KQL (Kusto Query Language) injection vulnerabilities in three MCP tool handlers: `get_table_schema`, `sample_table_data`, and `get_table_details`. The `table_name` parameter is interpolated directly into KQL queries via f-strings without any validation or sanitization, allowing an attacker (or a prompt-injected AI agent) to execute arbitrary KQL queries against the Azure Data Explorer cluster. Commit 0abe0ee55279e111281076393e5e966335fffd30 patches the issue.
CVSS 8.3
CVE-2026-34975 GITHUB HIGH python WORKING POC
Plunk <0.8.0 Raw MIME Construction - Email Header Injection
Plunk is an open-source email platform built on top of AWS SES. Prior to 0.8.0, a CRLF header injection vulnerability was discovered in SESService.ts, where user-supplied values for from.name, subject, custom header keys/values, and attachment filenames were interpolated directly into raw MIME messages without sanitization. An authenticated API user could inject arbitrary email headers (e.g. Bcc, Reply-To) by embedding carriage return/line feed characters in these fields, enabling silent email forwarding, reply redirection, or sender spoofing. The fix adds input validation at the schema level to reject any of these fields containing \r or \n characters, consistent with the existing validation already applied to the contentId field. This vulnerability is fixed in 0.8.0.
CVSS 8.5