Launching today
incursio
Security audits by AI agents, on your Claude or ChatGPT plan
1 follower
Security audits by AI agents, on your Claude or ChatGPT plan
1 follower
AI agents that audit your code for security vulnerabilities, on your Claude or ChatGPT subscription or on local models with Ollama. Based on Strix. Join the open beta.
Free
Launch Team
Dropbox SignSimple, secure eSignatures for the way your team works.
Promoted
Hey Product Hunt 👋
I’m Elias, and I built Incursio because security audits were the one part of shipping I kept putting off. Pentests are expensive and slow, scanners drown you in false positives, and pasting code into a chat window doesn’t scale
past a single file.
Incursio is a Mac app built on Strix, the open-source agent framework for security testing. Point it at a project folder and a team of AI agents works through your code the way a security reviewer would, then hands you a report of findings ranked by severity.
What makes it different:
🔑 No new bill. Run audits on the Claude or ChatGPT subscription you already pay for, on a local model with Ollama, or with your own Anthropic, OpenAI, Gemini or OpenRouter key.
🖥️ Runs on your Mac. Audits run locally in a Docker sandbox. Your code never goes to our servers, only to the model you choose (and nowhere at all with Ollama).
📊 Know where you stand. The dashboard shows open findings by severity, how fresh your last audit is, how many files changed since, and what the audit would have cost at API rates.
🔁 Audit only what changed. Re-run on just the files that changed since the last audit instead of starting over.
💸 Spending limits. On API keys, each audit stops at a limit you set ($10 by default). Usage rings show what’s left of your Claude and ChatGPT plans.
✅ No terminal needed. The first window sets up Docker, the sandbox and your model for you. Signed and notarized by Apple, and it updates itself.
It’s free during the open beta. Windows and Linux are on the way: there’s a waiting list on the site.
I’d love your feedback, especially on what it finds (and misses) in your own code. There’s a Send feedback button right in the app, and I’ll be here in the comments all day.