Agent Memory Guard: Prevent memory poisoning in DSPy agents with persistent state (OWASP) #9879
vgudur-dev
started this conversation in
Show and tell
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
What I built
Agent Memory Guard — an open-source Python middleware that screens memory reads/writes in AI agent systems for injection attacks, data poisoning, and exfiltration.
Why DSPy users should care
If you use DSPy's memory modules, conversation history, or any pipeline with persistent state across calls, your programs are vulnerable to memory poisoning — classified as OWASP ASI-06.
Attack scenario:
Integration with DSPy
Key features
Links
Would love feedback from the DSPy community — especially on the best integration points (custom module? assertion?).
Beta Was this translation helpful? Give feedback.
All reactions