AgentSec
Open source adversarial tests for AI agents: prompt injection, tool misuse, memory poisoning, and MCP scanning.
Open-source AI agents that run a full pentest end to end, from recon through exploit, reverse shell and pivot, then write a client-ready PDF report.
Open source adversarial tests for AI agents: prompt injection, tool misuse, memory poisoning, and MCP scanning.
Sandboxed AI harness in Zig. Landlock, eBPF, and namespaces isolate tool calls. WASM plugins over IPC.
Open-source CLI agent for authorized pentests. Plan, act, observe, verify, report. Local or hosted LLMs. Burp + MCP.
Open-source control plane for AI agents. Defense by default. Used by thousands of devs a month.
Your AI agent is probably less secure than you think. I built AgentSec to prove it. Open-source. Free. ⭐ Star it 🔁 Retweet so more builders see it 🧪 Use it. Break it. Find the flaws. 💬 Tell me what fails. Let's make AI agents safer together. github.com/invarislabs/in…
Just released a sandboxed AI harness. It's written purely in Zig and utilizes things such as landlock, eBPF, and namespaces to isolate tool calls for LLM's. It also utilizes WASM for plugins in an isolated process and fed back to the session over IPC. github.com/Midstall/chock
PentesterFlow: CLI agent for AI-assisted penetration testing PentesterFlow is an open-source CLI agent that follows a plan → act → observe → verify → report cycle. It supports local and hosted LLMs through Ollama, LM Studio, Gemini, Groq, and OpenAI-compatible APIs. The Show more
The ai agent control pattern is getting more validated, the reason we built prismor and used by 8k devs monthly github.com/PrismorSec/pri…