Yaniv Radunsky

REMnux OpenClaw skill for triage-first malware analysis: identifies the real payload, extracts high-value IOCs, hunts backend infrastructure, and produces concise markdown triage reports. #REMnux #OpenClaw #malware github.com/ionsec/remnux-…

LiteParse: fast, local PDF parser delivering spatial text extraction with bounding boxes, JSON/text outputs, built-in Tesseract.js OCR and pluggable OCR API. Screenshot generation for LLM agents. No cloud dependency. #LiteParse #PDF #OCR github.com/run-llama/lite…

Today we're announcing the Vibecode CLI Your OpenClaw (or any agent) can build and deploy up to 20 apps in parallel in a single prompt. Your agent will communicate with Vibecode to build your app(s). Frontend, Backend, Database, Deployments, APIs. Works with OpenClaw, Claude Code, Perplexity Computer, Manus, and any agent with access to a terminal. Just add this to your prompt: "Please use vibecode . dev CLI to build these apps"

LLM CLIs (eg. Gemini) can act as agentic operators in offensive tests—used with MCP servers to accelerate BloodHound AD analysis, triage Burp Suite traffic, and automate NTLM relay attack chains. #llm #bloodhound #pentesting armadin.com/blog-posts/aut…

NVIDIA NemoClaw (alpha) is an open-source reference stack that bundles the OpenShell runtime and OpenClaw agents with models like Nemotron for sandboxed always-on assistants. Early preview from Mar 16, 2026. #NemoClaw #OpenShell github.com/NVIDIA/NemoClaw

Machine identities now outnumber humans 82:1; attackers harvest machine creds (VoidLink, LangFlow, ShadowRay). SPIFFE/SPIRE provide short-lived workload identities but AI agents and static API keys remain an exposed gap. #machineidentity #SPIFFE #VoidLink blogs.cisco.com/security/ident…

We wanted to come on here to clear the air and confirm that the rumors are true... Cinematic Video Overviews are officially rolled out to 100% of Pro users in English! Please respect our privacy during this time by flooding our replies with your favorite creations.

Immich is a self-hosted photo & video manager for private backups, browsing, search and organization. Mobile apps for Android and iOS; source and localization on GitHub and Weblate. #selfhosted #privacy #Immich immich.app

RomM is a self-hosted ROM manager with built-in EmulatorJS, metadata from IGDB/Screenscraper/MobyGames, support for 400+ systems, AGPL-3.0 license, and device sync coming soon. #RomM #EmulatorJS #AGPL romm.app

AgentSonar maps process→domain pairs from outbound traffic, assigns an AI-likelihood score (0–1) using traffic-shape heuristics and known-agent rules, and surfaces high-scoring unknowns for triage. #AgentSonar #LLM #AI github.com/knostic/AgentS…

falcon-mcp is an MCP server connecting AI agents to CrowdStrike Falcon, exposing detections, incidents, and behavior data via modular APIs. Public preview; avoid production deployments. #falconMCP #CrowdStrike #MCP github.com/CrowdStrike/fa…

gstack: Garry Tan’s open-source "software factory" that turns Claude Code into a 15-role virtual engineering team; reported 600k LOC in 60 days and MIT license. #gstack #ClaudeCode #AI github.com/garrytan/gstack

Going to be spinning up a little series of posts with some Blue Team specific Claude tips - follow along with #ClaudeForBlueTeam First tip - ATT&CK can be downloaded as a JSON file, which Claude loves to work with. Remember, ATT&CK has a Data Source component - use Claude to ask some questions like "What logs do I need to detect X" or "What data source do I need to enable to cover the most techniques"

Data Israel: a multi-agent system that queries data.gov.il and CBS APIs live to answer Hebrew questions with sourced tables, charts, and transparent tool calls. #AI #OpenData #CBS github.com/LiorVainer/dat…

Agentic LLMs can orchestrate tools (strings, disasm, decompile) via MCP to automate iterative malware reverse engineering; structured pipelines improve coverage over standalone agents in a concrete case study. #LLM #malware #MCP synthesis.to/2026/03/18/age…

SOC Analyst Hub — Tier 1 provides 5 playbooks, decision flows for alert triage, structured hunting hypotheses, and a ~4-week learning path for Tier 1 analysts. #SOCTips #ThreatHunting #IR cross-samuel1.github.io/soc-analyst-hu…

Elastic Agent Skills: an official library of Agent Skills that teach AI coding agents to work with Elasticsearch, Kibana, Observability and Security. Includes SKILL.md packages and explicit security guidance. #elastic #agentskills github.com/elastic/agent-…

VMkatz extracts Windows secrets (NTLM, DPAPI keys, Kerberos tickets, LSA secrets, NTDS.dit) directly from VM snapshots and virtual disks on NAS/hypervisors. Single ~2.5MB binary; supports .vmsn, .vmdk, .sav. #VMkatz #vmdk #ntlm github.com/nikaiw/VMkatz

Genspark centralizes GPT-5.2, Claude 4.5 and Gemini 3 Pro into an agentic workspace with dedicated agents (AI Slides, AI Sheets, AI Developer), Mixture of Agents orchestration, Deep Research and integrations (Gmail/Notion/Slack). ~$20–25/mo #Genspark #AI brainai.co.il/guides/genspar…

Community repo: 611+ cybersecurity skills for AI agents using the agentskills.io standard. Compatible with Claude Code, GitHub Copilot, OpenAI Codex CLI, Gemini CLI and 20+ platforms. Apache-2.0. Community project. #agentskills_io #ClaudeCode #AI github.com/mukul975/Anthr…