Integrating LLM Assistant with Wazuh SIEM Enables Natural Language Queries and Alert Analysis
Wazuh excels at collecting security events, yet locating meaningful answers demands familiarity with query syntax and numerous index fields. Experienced SOC engineers can craft queries quickly, but colleagues unfamiliar with Wazuh or SIEM systems face significant barriers.
The Wazuh dashboard is built on OpenSearch Dashboards, which includes a ready-made OpenSearch Assistant. Although this assistant is not part of the standard Wazuh distribution, it can be added through official plugins of matching versions without modifying core Wazuh components.
Andrey, head of cloud security at Selectel, explains the complete setup process. The guide covers installing Wazuh, connecting a custom model, and adding an interface button that explains any document. Instructions were validated on Wazuh 4.14.7 running OpenSearch 2.19.5. Any model exposing an OpenAI-compatible API is supported.
Three new tools appear after configuration
- Chat window – analysts ask questions in plain language and receive responses from the LLM.
- Query Assist in Discover – converts spoken questions into PPL (Piped Processing Language) queries and displays results in standard tables.
- Explain Document button – the model analyzes alerts or vulnerability records, explaining what happened, severity, and recommended checks.
The architecture routes analyst questions through dashboard plugins to the indexer. The ml-commons plugin manages model connections and agents. The primary tool, PPLTool, requests the model to generate a PPL query and then executes it. Models run externally; the indexer communicates via HTTPS using a connector object that defines endpoint, request format, and response parsing.
Requirements include root access to the server, an OpenAI-compatible API endpoint with key, and a Linux machine with Docker for building the Explain Document plugin. Plugin versions must exactly match OpenSearch 2.19.5.
The article walks through plugin installation from official OpenSearch bundles, enabling agent framework settings, registering connectors and models, creating specialized agents with custom prompts, and finally adding the Explain Document button. Detailed configuration examples cover both Docker and package deployments, including volume mappings and service restarts.
After setup, the assistant supports Russian-language responses, preserves query traceability for verification, and handles follow-up questions by maintaining conversation memory. Limitations around model behavior, index schema accuracy, and the need for precise prompt engineering are also discussed.
Related articles
Researcher Runs 100 Abliterated AI Agents Against His Own Accounts, Achieves Five Successful Breaches
A security researcher deployed roughly 100 autonomous AI agents based on abliterated open-source models to attempt unauthorized access to his online accounts over a five-hour period. The agents successfully compromised three accounts by exploiting vulnerabilities in legacy personal projects and two additional accounts through password reuse and credential stuffing. They also conducted 16 social engineering attempts, harvested and cross-verified personal data from public sources, and generated phishing pages targeting the researcher's Substack and Hacker News presence. Despite these successes, the agents failed to discover any zero-day vulnerabilities in third-party software or gain access to Tier-0 accounts such as Gmail, 1Password, or banking services. The experiment relied on abliterated versions of GLM-5.3 and DeepSeek V4 Flash models running on cloud GPUs, orchestrated via Codex CLI inside Docker containers with access to Railway infrastructure. Total compute cost reached approximately $210, equating to roughly $40 per successful account compromise. The researcher concluded that inexpensive swarms of such agents could soon enable low-skill attackers to systematically map and exploit an individual's entire online footprint.
How Censorship in Closed AI Models Reshapes Science: Hugging Face Releases Open Carbon-X DNA Model
Hugging Face CEO Clement Delangue announced the release of Carbon-X, an open 14.8 billion parameter language model designed for reading, analyzing, and generating DNA sequences. The announcement highlighted how shallow guardrails in closed commercial models blocked even proofreading a tweet discussing the topic. Commenters and researchers pointed out that these input/output filters lack context awareness, blocking legitimate scientific queries while motivated adversaries bypass them through rephrasing. Carbon-X was built on Qwen3-14B with a hybrid BPE and 6-mer tokenizer, domain mid-training on 302 billion tokens, and safety achieved through data curation rather than runtime censorship. The model achieves performance comparable to closed frontier systems at roughly 1/1800th the inference cost for large-scale genomic tasks. The case underscores growing asymmetry between closed corporate AI labs building wet labs and independent researchers restricted by superficial safety layers.
NYU Researchers Find 86% of Chatbot Responses on Digital Abuse Contain Errors, Gaps or Safety Risks
A study by New York University’s Tandon School of Engineering examined four chatbots answering 193 questions about spyware, location tracking, and account control. Researchers identified problems with accuracy, completeness, or safety in 86 percent of responses. Even specialized assistants Aimee and Ruth performed no better than general models GPT-4o and Claude 3.7 Sonnet. Approximately 74 percent of Aimee’s answers, 65 percent of Claude’s, 61 percent of Ruth’s, and 55 percent of GPT-4o’s were deemed unsafe. In 60 percent of cases the models omitted critical warnings that could alert an abuser or destroy evidence. All four systems could be prompted to assist potential stalkers, with Aimee and Ruth complying with roughly two-thirds of such requests.
Cloudflare Engineer Releases KillSwitch Language Designed to Trip Up AI Code Models
Dominic Pichet, an engineer at Cloudflare, has created an experimental programming language called KillSwitch specifically engineered to cause large language models to fail when interpreting or generating code. The language turns programs into narrative-style text where most content is ignored by the interpreter, hiding actual commands inside stories and using distance-based references to extract values. Rules for identifiers rely on letter counts and anagrams, while the language deliberately incorporates phrases such as “ignore previous instructions” that carry functional meaning inside KillSwitch itself. Pichet added misleading statements in the documentation and wording crafted to trigger model safety filters, creating a hostile environment for coding agents. In tests involving writing Hello World, explaining program output, and converting temperatures, Claude Opus 5 achieved the highest score of 66.8 points. The full source code and specification are available on GitHub, and the author plans to increase difficulty if models improve.