AgentInterdict

Your agent is one document away from a breach.

5 followers

Your agent is one document away from a breach.

5 followers

AgentInterdict is runtime enforcement for autonomous AI agents — interdicting unsafe action by verifying authority, tracking provenance, and revalidating at action time.

Free

Launch tags:Open Source•Privacy•Artificial Intelligence

Launch Team

ClarifyThe Full-Stack AI CRM that replaces 5 tools

Promoted

Maker

📌

It started with an uncomfortable realization: I was giving my AI agents more memory, more tools, and more autonomy — while almost nobody was securing the boundary where context becomes action. I kept seeing the same gap in every agent framework. A single poisoned document, a single malicious email, a single compromised web page could plant an instruction my agent would obey without question. My API keys, my credentials, my agent's integrity — all one bad retrieval away from being compromised. The uncomfortable truth that pushed me over the edge: if you run an agent with memory and tools and no enforcement layer, you are one malicious document away from a breach. Not "maybe." One document. So I built the thing I couldn't find: a local, open-source runtime that interdicts unsafe action before it executes — binds authority to origin, blocks credential leakage, quarantines memory poisoning, and revalidates every action at action-time. And I made a deliberate choice about honesty. Most security tools claim 99.9% block rates. I ran 200 real injection attempts through mine and got 96.5% — and I published the 7 misses with their payloads, because a security tool that hides its misses isn't trustworthy. It's free for personal/research use. The core that keeps your agent safe shouldn't be a paywall. → [github.com/BryanFiFife/AgentInterdict] What attack vectors am I missing? I'd genuinely love the community's input.

Report

3mo ago

Forum Threads

p/agentinterdict

•

3mo ago

What attack vectors should AgentInterdict cover next?

I built AgentInterdict to secure the boundary where an AI agent's context becomes action stopping prompt injection, credential leakage, memory poisoning, and authority spoofing before they can cause a breach.
I benchmarked it against 200 real injection attempts and published the results honestly: 96.5% block rate, 7 documented misses (I publish the misses rather than claim 99.9% a security tool that hides its failures isn't trustworthy).
The 7 misses are the gaps I care about most. They break down roughly as:
- multi-turn / conversational injection that spans several exchanges
- obfuscation techniques I haven't seen yet
- edge cases in tool-call semantics
So here's the question for this thread: what attack vectors or evasion techniques would you want an agent-security runtime to handle next?
Whether you're building agents, running local LLMs, or thinking about this from a security perspective I want the hard cases. What am I missing?
Open-source, local-first, free for personal/research use: github.com/BryanFiFife/AgentInterdict

View all