#1LLM security, prompt injection, model extraction, adversarial AI, and AI red teaming tools.
Kitploit recommended

A black-box optimization approach for protecting Large Language Model (LLM) system prompts against adversarial extraction attacks.
Is 4-bit quantization more than noise to a chat model's safety behaviour? Code, pre-registration and labels.

🐾 Authority, not credentials, for AI agents: policy-approved actions, delegation that can only narrow, instant revocation, tamper-evident audit.

A Loom registry admin can rewrite a skill document that has no group owner. The next full deploy of an agent that already lists that skill copies the…

On Loom, an approval rule named code_interpreter pauses the Strands tool and misses the ADK code and command tools. Empty, star, and…

On Loom with google-adk 2.13.0, an agent-to-agent tool never waits for the approval check, so the send completes. The same rule still pauses Strands…

High-performance open-source security scanner combining SAST, SCA, Secret Detection, and IaC analysis, built for developers and CI/CD pipelines,…

MCP server exposing VirusTotal threat intelligence to AI agents: file, URL, domain and IP reports, analysis submission, and receipt recovery via VTAI.

Make your own app expensive to reverse. A build-time audit, harden and verify harness for native binaries and mobile apps, built to counter…

Framework for training LLM cybersecurity agents without runtime environments, synthesizing CTF solving trajectories via persona-driven simulation and…

RuralShield — an AWS CDK research prototype for healthcare cloud security, with policy-bounded AI triage, human-approved response, audit logging, and…

Fast, AI-first web security testing toolkit in Rust: an open-source alternative to ZAP, Burp Suite and the like. Intercepting proxy, scanner, fuzzer…

Self-hosted OSINT platform. Point it at a domain, IP, email, phone or username and 26 modules run in parallel: WHOIS, DNS, threat intel, breaches,…

Self-hosted OSINT platform. Point it at a domain, IP, email, phone or username and 26 modules run in parallel: WHOIS, DNS, threat intel, breaches,…

Let AI agents explore a running game from the inside. Live, safe, instant.

Self-service pentest platform that only scans domains whose ownership is proven, and re-checks that proof at scan time. OWASP, TLS, DNS and port…

AI multi-agent autonomous penetration testing platform that runs recon, attack-surface discovery, exploitation validation and evidence-backed report…

Benchmark harness that evaluates LLM agent tool-calling security under long-horizon tasks, generating prompt-injection and tool-poisoning attacks and…