2026-11-14
15 categories analyzed.
Malware Analysis
- dagoSte/reverseai — 19 stars — Static reverse engineering of untrusted files with verifiable, byte-anchored evidence. Open tools (Ghidra, JADX, LIEF, YARA-X, Babel) in locked offline containers, from the command line or from AI agents over MCP.
- gitbilelx/katyusha-v32-writeup — 2 stars — Static analysis write-up, IOCs, and YARA rules for the Katyusha/v32 PowerBASIC RAT/proxy-bot kit (defensive research)
- whispering3/Sentinel-PE — 2 stars — Local-first Windows PE malware triage workbench & endpoint guard. Explainable static analysis, ransomware/infostealer heuristics, YARA/Sigma generator, and AES-256-GCM DPAPI quarantine vault.
SOC Automation
- xsignon-secure/Chimera — 3 stars — Chimera: an MCP honeypot that catches rogue AI agents, hands them tracked fake secrets, and lets your Blue Team block them in one click. Free and open source.
- akquisio-dev/vigil — 2 stars — Lightweight SOC for small Linux server fleets: signed agents, canaries, tamper-evident event log, MITRE ATT&CK-mapped alerts. Zero dependencies.
- dururustem68/wazuh-daily-triage — 2 stars — AI-assisted daily triage for Wazuh SIEM
Incident Response
- Pager-Who/PagerWho — 3 stars — Open-source, self-hosted incident response with on-call schedules, escalation policies, webhooks, and notifications.
- banuchnik-byte/enterprise-ir-investigation — 2 stars — Simulated enterprise incident investigation covering phishing analysis, memory forensics, DNS exfiltration, and SIEM correlation, with MITRE ATT&CK mapping.
- vivekyarra/Graph — 2 stars — MAELIN: graph-native incident response on FalkorDB. Agents find every attack path, prove the minimal cut (algo.maxFlow), replay it on a graph clone, and remember outcomes across sessions. Graph Hacks 2026.
Compliance / GRC
few valid results today
- VenusOne-Lee/popia-compliance — 12 stars — Open-source POPIA privacy and security guardrails for South African builders, for every AI coding tool.
- Nimstead/assure — 2 stars — Find what’s vulnerable, fix it safely with automatic rollback, and prove it to an auditor. Self-hosted vulnerability management for Linux, Windows and SBOMs.
AI for Cybersecurity
- SpecterOps/Jailbreaker-CE — 118 stars — Jailbreaker is a local evaluation tool for testing chatbot and agent-style systems against jailbreak, prompt-injection, and related failure modes.
- Attention-Span-AI/Tego — 24 stars — Tego limits what any AI agent can touch on Linux, and records every run. Rules for MCP servers, CI jobs, Python workers, 24/7 assistants and coding agents, enforced by the kernel.
- guard-spec/guard-spec — 15 stars — Self-hosted prompt-injection firewall for LLM apps. OpenAI-compatible gateway with rules, an ONNX classifier and embeddings, output guard, alerts, SSO. English and Vietnamese.
- rdwr-gabin/github-ipi-in-the-wild — 8 stars — GitHub-scale measurement of indirect prompt injection of coding agents in the wild
Reverse Engineering
- aleph-void/Ghidra-Rare — 25 stars — A repository of SLEIGH support for rare processor ISAs
- muhammadrizwan87/frida-hook-engine — 16 stars — Config-driven Frida instrumentation engine for Android security researchers and reverse engineers. Zero scripting required, built for simplicity and ease of use.
- AlexDev-TH/ECU-ArchID — 11 stars — Detect the CPU family of ECU firmware dumps (.bin .hex .s19 .elf) from instruction patterns, with measured confidence. V850, RH850, SuperH, M16C, R8C, PowerPC, TriCore, HCS12, H8, RL78 and more.
- HappyTotoro116/ghidra-llm-pipeline — 11 stars — Deterministic PyGhidra extraction, triage, and local LLM compiler-backend pipeline for C reimplementation.
Fuzzing
- MozillaSecurity/funfuzz — 645 stars — A collection of fuzzers in a harness for testing the SpiderMonkey JavaScript engine.
- keldxd/invariant-vacuity — 2 stars — Find Foundry invariant tests that pass for the wrong reason.
- p17433308-cmyk/SOMEIPwn — 2 stars — First offensive security framework for Automotive Ethernet (SOME/IP & DoIP) — service discovery, MITM proxy, UDS security-access cracking, grammar-aware fuzzer with live crash detection
Threat Intelligence
- oasis-open/cti-stix-common-objects — 104 stars — OASIS Cyber Threat Intelligence (CTI) TC: A repository for commonly used STIX objects in order to avoid needless duplication. https://github.com/oasis-open/cti-stix-common-objects
- LeenAdeeb/IOC-Enricher — 2 stars — Python CLI that enriches IOCs (VirusTotal, AbuseIPDB, Shodan, OTX, abuse.ch), maps infrastructure with graph pivoting, and ranks possible APT groups using MITRE ATT&CK and MISP Galaxy.
- mhearsh/threat-intel-feed — 2 stars — Curated high-confidence IOC feed (IP, domain, URL, file hashes), 0-100 scored, every 2h. Ready-to-import for Palo Alto, FortiGate, Check Point, Cisco, Defender, CrowdStrike, SentinelOne, Splunk, Sentinel, STIX 2.1.
- ThreatConnect-Inc/skills — 2 stars — Dataminr CTI skills for Claude, Codex and Copilot: threat intelligence management, playbook and agent building, and automation. Claude Code plugin marketplace.
Detection Engineering / SIEM
- kiddsaut07/detectforge — 2 stars — Detection engineering toolkit for parsing, processing, validating and converting DetectForge rules into SIEM and query languages
- qays3/Detection-Engineering-Range — 2 stars — One home lab that trains anyone to become a detection engineer. Every tool is free. One core server runs the platform in Docker, three virtual machines play attacker and victims, and one setup script per machine builds everything.
- Xusniddin-Security/ad-siem-detector — 2 stars — Catch Active Directory attackers in real time. Agent + backend + live dashboard that detects Kerberoasting, DCSync, password spraying & 10 more attacks, mapped to MITRE ATT&CK, with instant Telegram alerts. Pure Python, zero dependencies.
Penetration Testing
few valid results today
- fzn0x/watchtower — 222 stars — Watchtower - Beginner-friendly AI-powered penetration testing automation CLI with orchestrated agentic workflows.
- shencah/autorecon — 12 stars — Universal frontend endpoint recon tool - crawls JS, extracts APIs, probes them, highlights login/crypto logic
Cloud Security / DevSecOps
- Hero-all/repo-airgap — 8 stars — 帮你查出开源软件“偷偷联网”的地方(上报数据、偷偷下载、弹广告),让它能在断网的内部网络里放心使用。Find where open-source software quietly goes online, so it can run safely offline.
- hugomn/heldkeys — 8 stars — Which keys is your agent holding? Lists the credentials an AI agent can reach on your laptop, in a sandbox, or in CI, and flags risky AI agent steps in GitHub workflows.
- Flurit/applygate — 6 stars — A check between AI coding agents and terraform apply. It reads the plan, not just the command, so an apply that would replace a prod database is stopped while safe ones go through. Local allow/ask/block hook for Claude Code and Codex; preview hooks for Cursor, Copilot CLI, Gemini CLI, Windsurf, Devin, OpenCode and Cline. Terraform and OpenTofu.
- HuginnsHammer/CloudSentry — 4 stars — Lightweight static security auditor for Terraform, Dockerfiles and Docker Compose. Detect misconfigurations, assess risk and get actionable remediation guidance.
Vulnerability Research
- sanjio3/SAVR — 5 stars — Automated white-box vulnerability scanning and verification: SAVR confirms every SAST finding by deterministically replaying it against a live target. Scan · Analyze · Verify · Replay
- ljw097/REMNANT — 2 stars — REMNANT is a automated historical vulnerability research engine.
- TypeThe0ry/security-advisories — 2 stars — Coordinated security advisories and vulnerability research by TypeThe0ry
Red Team / C2
few valid results today
- 0x7AK7/RedVector — 40 stars — RedVector AI - AI-powered penetration testing and cybersecurity automation framework for security researchers and ethical hackers.
OSINT
- The-Osint-Toolbox/Telegram-OSINT — 2.1k stars — In-depth repository of Telegram OSINT resources covering, tools, techniques & tradecraft.
- The-Osint-Toolbox/Social-Media-OSINT — 1.1k stars — Social Media OSINT collection containing - tools, techniques & tradecraft.
- The-Osint-Toolbox/Image-Research-OSINT — 577 stars — Learn how to research images and the tools, techniques & tradecraft required.
DFIR / Forensics
few valid results today
- jankammerath/BogusMips — 8 stars — Forensic analytics for disassembly, reassembly, reverse engineering and auditing of Android apps
- kalink0/4n6-notes — 2 stars — Digital forensic artifacts at a glance, built from ten years of collected notes: what they are, what they prove, where to find them, with references and free tools. Windows, macOS, iOS, Android, Linux and more.
Top picks today
- xsignon-secure/Chimera
- A deception layer built for the MCP era: fake secrets that track rogue AI agents for blue-team blocking.
- SpecterOps/Jailbreaker-CE
- SpecterOps releases a local, repeatable harness for red-teaming chatbots and agents against jailbreak and prompt injection.
- kalink0/4n6-notes
- A decade of forensic-artifact notes across Windows, macOS, iOS, Android and Linux in one quick reference.
2026-11-14
15 categorie analizzate.
Malware Analysis
- dagoSte/reverseai — 19 stelle — Static reverse engineering of untrusted files with verifiable, byte-anchored evidence. Open tools (Ghidra, JADX, LIEF, YARA-X, Babel) in locked offline containers, from the command line or from AI agents over MCP.
- gitbilelx/katyusha-v32-writeup — 2 stelle — Static analysis write-up, IOCs, and YARA rules for the Katyusha/v32 PowerBASIC RAT/proxy-bot kit (defensive research)
- whispering3/Sentinel-PE — 2 stelle — Local-first Windows PE malware triage workbench & endpoint guard. Explainable static analysis, ransomware/infostealer heuristics, YARA/Sigma generator, and AES-256-GCM DPAPI quarantine vault.
SOC Automation
- xsignon-secure/Chimera — 3 stelle — Chimera: an MCP honeypot that catches rogue AI agents, hands them tracked fake secrets, and lets your Blue Team block them in one click. Free and open source.
- akquisio-dev/vigil — 2 stelle — Lightweight SOC for small Linux server fleets: signed agents, canaries, tamper-evident event log, MITRE ATT&CK-mapped alerts. Zero dependencies.
- dururustem68/wazuh-daily-triage — 2 stelle — AI-assisted daily triage for Wazuh SIEM
Incident Response
- Pager-Who/PagerWho — 3 stelle — Open-source, self-hosted incident response with on-call schedules, escalation policies, webhooks, and notifications.
- banuchnik-byte/enterprise-ir-investigation — 2 stelle — Simulated enterprise incident investigation covering phishing analysis, memory forensics, DNS exfiltration, and SIEM correlation, with MITRE ATT&CK mapping.
- vivekyarra/Graph — 2 stelle — MAELIN: graph-native incident response on FalkorDB. Agents find every attack path, prove the minimal cut (algo.maxFlow), replay it on a graph clone, and remember outcomes across sessions. Graph Hacks 2026.
Compliance / GRC
pochi risultati validi oggi
- VenusOne-Lee/popia-compliance — 12 stelle — Open-source POPIA privacy and security guardrails for South African builders, for every AI coding tool.
- Nimstead/assure — 2 stelle — Find what’s vulnerable, fix it safely with automatic rollback, and prove it to an auditor. Self-hosted vulnerability management for Linux, Windows and SBOMs.
AI for Cybersecurity
- SpecterOps/Jailbreaker-CE — 118 stelle — Jailbreaker is a local evaluation tool for testing chatbot and agent-style systems against jailbreak, prompt-injection, and related failure modes.
- Attention-Span-AI/Tego — 24 stelle — Tego limits what any AI agent can touch on Linux, and records every run. Rules for MCP servers, CI jobs, Python workers, 24/7 assistants and coding agents, enforced by the kernel.
- guard-spec/guard-spec — 15 stelle — Self-hosted prompt-injection firewall for LLM apps. OpenAI-compatible gateway with rules, an ONNX classifier and embeddings, output guard, alerts, SSO. English and Vietnamese.
- rdwr-gabin/github-ipi-in-the-wild — 8 stelle — GitHub-scale measurement of indirect prompt injection of coding agents in the wild
Reverse Engineering
- aleph-void/Ghidra-Rare — 25 stelle — A repository of SLEIGH support for rare processor ISAs
- muhammadrizwan87/frida-hook-engine — 16 stelle — Config-driven Frida instrumentation engine for Android security researchers and reverse engineers. Zero scripting required, built for simplicity and ease of use.
- AlexDev-TH/ECU-ArchID — 11 stelle — Detect the CPU family of ECU firmware dumps (.bin .hex .s19 .elf) from instruction patterns, with measured confidence. V850, RH850, SuperH, M16C, R8C, PowerPC, TriCore, HCS12, H8, RL78 and more.
- HappyTotoro116/ghidra-llm-pipeline — 11 stelle — Deterministic PyGhidra extraction, triage, and local LLM compiler-backend pipeline for C reimplementation.
Fuzzing
- MozillaSecurity/funfuzz — 645 stelle — A collection of fuzzers in a harness for testing the SpiderMonkey JavaScript engine.
- keldxd/invariant-vacuity — 2 stelle — Find Foundry invariant tests that pass for the wrong reason.
- p17433308-cmyk/SOMEIPwn — 2 stelle — First offensive security framework for Automotive Ethernet (SOME/IP & DoIP) — service discovery, MITM proxy, UDS security-access cracking, grammar-aware fuzzer with live crash detection
Threat Intelligence
- oasis-open/cti-stix-common-objects — 104 stelle — OASIS Cyber Threat Intelligence (CTI) TC: A repository for commonly used STIX objects in order to avoid needless duplication. https://github.com/oasis-open/cti-stix-common-objects
- LeenAdeeb/IOC-Enricher — 2 stelle — Python CLI that enriches IOCs (VirusTotal, AbuseIPDB, Shodan, OTX, abuse.ch), maps infrastructure with graph pivoting, and ranks possible APT groups using MITRE ATT&CK and MISP Galaxy.
- mhearsh/threat-intel-feed — 2 stelle — Curated high-confidence IOC feed (IP, domain, URL, file hashes), 0-100 scored, every 2h. Ready-to-import for Palo Alto, FortiGate, Check Point, Cisco, Defender, CrowdStrike, SentinelOne, Splunk, Sentinel, STIX 2.1.
- ThreatConnect-Inc/skills — 2 stelle — Dataminr CTI skills for Claude, Codex and Copilot: threat intelligence management, playbook and agent building, and automation. Claude Code plugin marketplace.
Detection Engineering / SIEM
- kiddsaut07/detectforge — 2 stelle — Detection engineering toolkit for parsing, processing, validating and converting DetectForge rules into SIEM and query languages
- qays3/Detection-Engineering-Range — 2 stelle — One home lab that trains anyone to become a detection engineer. Every tool is free. One core server runs the platform in Docker, three virtual machines play attacker and victims, and one setup script per machine builds everything.
- Xusniddin-Security/ad-siem-detector — 2 stelle — Catch Active Directory attackers in real time. Agent + backend + live dashboard that detects Kerberoasting, DCSync, password spraying & 10 more attacks, mapped to MITRE ATT&CK, with instant Telegram alerts. Pure Python, zero dependencies.
Penetration Testing
pochi risultati validi oggi
- fzn0x/watchtower — 222 stelle — Watchtower - Beginner-friendly AI-powered penetration testing automation CLI with orchestrated agentic workflows.
- shencah/autorecon — 12 stelle — Universal frontend endpoint recon tool - crawls JS, extracts APIs, probes them, highlights login/crypto logic
Cloud Security / DevSecOps
- Hero-all/repo-airgap — 8 stelle — 帮你查出开源软件“偷偷联网”的地方(上报数据、偷偷下载、弹广告),让它能在断网的内部网络里放心使用。Find where open-source software quietly goes online, so it can run safely offline.
- hugomn/heldkeys — 8 stelle — Which keys is your agent holding? Lists the credentials an AI agent can reach on your laptop, in a sandbox, or in CI, and flags risky AI agent steps in GitHub workflows.
- Flurit/applygate — 6 stelle — A check between AI coding agents and terraform apply. It reads the plan, not just the command, so an apply that would replace a prod database is stopped while safe ones go through. Local allow/ask/block hook for Claude Code and Codex; preview hooks for Cursor, Copilot CLI, Gemini CLI, Windsurf, Devin, OpenCode and Cline. Terraform and OpenTofu.
- HuginnsHammer/CloudSentry — 4 stelle — Lightweight static security auditor for Terraform, Dockerfiles and Docker Compose. Detect misconfigurations, assess risk and get actionable remediation guidance.
Vulnerability Research
- sanjio3/SAVR — 5 stelle — Automated white-box vulnerability scanning and verification: SAVR confirms every SAST finding by deterministically replaying it against a live target. Scan · Analyze · Verify · Replay
- ljw097/REMNANT — 2 stelle — REMNANT is a automated historical vulnerability research engine.
- TypeThe0ry/security-advisories — 2 stelle — Coordinated security advisories and vulnerability research by TypeThe0ry
Red Team / C2
pochi risultati validi oggi
- 0x7AK7/RedVector — 40 stelle — RedVector AI - AI-powered penetration testing and cybersecurity automation framework for security researchers and ethical hackers.
OSINT
- The-Osint-Toolbox/Telegram-OSINT — 2.1k stelle — In-depth repository of Telegram OSINT resources covering, tools, techniques & tradecraft.
- The-Osint-Toolbox/Social-Media-OSINT — 1.1k stelle — Social Media OSINT collection containing - tools, techniques & tradecraft.
- The-Osint-Toolbox/Image-Research-OSINT — 577 stelle — Learn how to research images and the tools, techniques & tradecraft required.
DFIR / Forensics
pochi risultati validi oggi
- jankammerath/BogusMips — 8 stelle — Forensic analytics for disassembly, reassembly, reverse engineering and auditing of Android apps
- kalink0/4n6-notes — 2 stelle — Digital forensic artifacts at a glance, built from ten years of collected notes: what they are, what they prove, where to find them, with references and free tools. Windows, macOS, iOS, Android, Linux and more.
Più interessanti oggi
- xsignon-secure/Chimera
- Un livello di deception pensato per l’era MCP: segreti finti che tracciano gli agenti AI ostili, utili al blocco da parte del blue team.
- SpecterOps/Jailbreaker-CE
- SpecterOps pubblica un harness locale e ripetibile per mettere alla prova chatbot e agenti contro jailbreak e prompt injection.
- kalink0/4n6-notes
- Dieci anni di note sugli artifact forensi, su Windows, macOS, iOS, Android e Linux, in un unico riferimento rapido.