Security Vulnerabilities
LlamaIndex (run-llama/llama_index) versions up to and including 0.11.6 contain an unsafe deserialization vulnerability in BGEM3Index.load_from_disk() in llama_index/indices/managed/bge_m3/base.py. The function uses pickle.load() to deserialize multi_embed_store.pkl from a user-supplied persist_dir without validation. An attacker who can provide a crafted persist directory containing a malicious pickle file can trigger arbitrary code execution when the victim loads the index from disk.
CVSS Score
7.8
EPSS Score
0.001
Published
2026-01-12
LlamaIndex (run-llama/llama_index) versions up to and including 0.12.2 contain an uncontrolled resource consumption vulnerability in the VannaPack VannaQueryEngine implementation. The custom_query() logic generates SQL statements from a user-supplied prompt and executes them via vn.run_sql() without enforcing query execution limits In downstream deployments where untrusted users can supply prompts, an attacker can trigger expensive or unbounded SQL operations that exhaust CPU or memory resources, resulting in a denial-of-service condition. The vulnerable execution path occurs in llama_index/packs/vanna/base.py within custom_query().
CVSS Score
7.5
EPSS Score
0.001
Published
2026-01-12
LangChain versions up to and including 0.3.1 contain a regular expression denial-of-service (ReDoS) vulnerability in the MRKLOutputParser.parse() method (libs/langchain/langchain/agents/mrkl/output_parser.py). The parser applies a backtracking-prone regular expression when extracting tool actions from model output. An attacker who can supply or influence the parsed text (for example via prompt injection in downstream applications that pass LLM output directly into MRKLOutputParser.parse()) can trigger excessive CPU consumption by providing a crafted payload, causing significant parsing delays and a denial-of-service condition.
CVSS Score
7.5
EPSS Score
0.001
Published
2026-01-12
Ollama 0.11.5-rc0 through current version 0.13.5 contain a null pointer dereference vulnerability in the multi-modal model image processing functionality. When processing base64-encoded image data via the /api/chat endpoint, the application fails to validate that the decoded data represents valid media before passing it to the mtmd_helper_bitmap_init_from_buf function. This function can return NULL for malformed input, but the code does not check this return value before dereferencing the pointer in subsequent operations. A remote attacker can exploit this by sending specially crafted base64 image data that decodes to invalid media, causing a segmentation fault and crashing the runner process. This results in a denial of service condition where the model becomes unavailable to all users until the service is restarted.
CVSS Score
7.5
EPSS Score
0.002
Published
2026-01-12
WebErpMesv2 is a Resource Management and Manufacturing execution system Web for industry. Prior to 1.19, the WebErpMesV2 application exposes multiple sensitive API endpoints without authentication middleware. An unauthenticated remote attacker can read business-critical data including companies, quotes, orders, tasks, and whiteboards. Limited write access allows creation of company records and full manipulation of collaboration whiteboards. This vulnerability is fixed in 1.19.
CVSS Score
8.2
EPSS Score
0.001
Published
2026-01-12
WebErpMesv2 is a Resource Management and Manufacturing execution system Web for industry. Prior to 1.19, WebErpMesv2 contains a file upload validation bypass vulnerability in multiple controllers that allows authenticated users to upload arbitrary files, including PHP scripts, leading to Remote Code Execution (RCE). This vulnerability is identical in nature to CVE-2025-52130 but exists in different code locations that were not addressed by the original fix. This vulnerability is fixed in 1.19.
CVSS Score
5.4
EPSS Score
0.0
Published
2026-01-12
Appsmith is a platform to build admin panels, internal tools, and dashboards. Prior to 1.93, the server uses the Origin value from the request headers as the email link baseUrl without validation. If an attacker controls the Origin, password reset / email verification links in emails can be generated pointing to the attacker’s domain, causing authentication tokens to be exposed and potentially leading to account takeover. This vulnerability is fixed in 1.93.
CVSS Score
9.6
EPSS Score
0.0
Published
2026-01-12
Emlog is an open source website building system. emlog v2.6.1 and earlier exposes a REST API endpoint (/index.php?rest-api=upload) for media file uploads. The endpoint fails to implement proper validation of file types, extensions, and content, allowing authenticated attackers (with a valid API key or admin session cookie) to upload arbitrary files (including malicious PHP scripts) to the server. An attacker can obtain the API key either by gaining administrator access to enable the REST API setting, or via information disclosure vulnerabilities in the application. Once uploaded, the malicious PHP file can be executed to gain remote code execution (RCE) on the target server, leading to full server compromise.
CVSS Score
8.8
EPSS Score
0.004
Published
2026-01-12
A SQL Injection was found in the /exam/user/profile.php page of kashipara Online Exam System V1.0, which allows remote attackers to execute arbitrary SQL command to get unauthorized database access via the rname, rcollage, rnumber, rgender and rpassword parameters in a POST HTTP request.
CVSS Score
9.1
EPSS Score
0.0
Published
2026-01-12
Lychee is a free, open-source photo-management tool. Prior to 7.1.0, an authorization vulnerability exists in Lychee's album password unlock functionality that allows users to gain possibly unauthorized access to other users' password-protected albums. When a user unlocks a password-protected public album, the system automatically unlocks ALL other public albums that share the same password, resulting in a complete authorization bypass. This vulnerability is fixed in 7.1.0.
CVSS Score
4.3
EPSS Score
0.0
Published
2026-01-12