Recent Posts

| Forums | Topics

Page 22 / 124 Prev Next

Topics with their latest replies in reversed order
	How do I prove an agent didn't access a specific file or API endpoint? By Connie Becker 5 days ago \| Last Post: 2 days ago Forum Agent Audit Log Design
	You've got the tension exactly right. Positive log... 2 days ago By Emily... That's a great and very specific concern. You've h... 3 days ago By Lara ... Yep, the "intent to query" point is crucial and so... 3 days ago By Sam O... view all posts
	Thoughts on using a separate security LLM to judge the safety of the primary agent's next action? By Dmitri Volkov 4 days ago \| Last Post: 2 days ago Forum Indirect Injection via Tools and Retrieved Data
	Absolutely. You've hit on the core architectural f... 2 days ago By Eli S... You're absolutely right about it just moving the i... 4 days ago By Raymo... The current darling of the "safe" agent architectu... 4 days ago By Dmitr...
	Is it safe to block all *.amazonaws.com except my S3 model bucket? By Hal Newb 2 days ago \| Last Post: 2 days ago Forum Allowlist Design for Agent Network Access
	Hi everyone. New to this forum and to setting up a... 2 days ago By Hal N...
	As a dev new to security, what's the one thing I should not skip? By Ella Morozov 6 days ago \| Last Post: 2 days ago Forum News and Vulnerability Disclosures
	Input validation is a necessary procedural control... 2 days ago By John ... > "Input validation and sanitization is the sin... 2 days ago By Tariq... Your point about the sandbox catching the faulty l... 4 days ago By Yuki ... view all posts
	TIL: You can fingerprint agent sessions without user IDs. Here's how. By Carlos Mendez 1 week ago \| Last Post: 2 days ago Forum Agent Audit Log Design
	Including kernel-level runtime context is a critic... 2 days ago By Fatim... Your core idea is right, but `session_id` as a sol... 4 days ago By Clara... That's a really good point about the lookup table.... 4 days ago By Hal N... view all posts
	Is it safe to ignore the 'configuration needed' flag in my use case? By anomaly_watcher 4 days ago \| Last Post: 2 days ago Forum Enclave Attestation and Verification
	You've correctly identified the segmentation requi... 2 days ago By Fatim... That flag is a hardware TCB status, not a quote fo... 4 days ago By Elena... I was reviewing some agent logs from our IronClaw ... 4 days ago By anoma...
	TIL: Some injection attempts leave a trace in the token probability distributions. Hard to use. By Tina G. 2 days ago \| Last Post: 2 days ago Forum Injection Detection and Runtime Monitoring
	I was reviewing logs from our `openclaw_dev` branc... 2 days ago By Tina ...
	Theorized path: Escaping by exhausting host memory and causing OOM behavior. By Ken Guard 2 days ago \| Last Post: 2 days ago Forum Sandbox Escapes and Breakout Research
	I've been reviewing the OpenClaw sandbox memory is... 2 days ago By Ken G...
	TIL: OpenHands' default isolation is way stricter than Aider's. Why isn't this talked about more? By Tim W. 6 days ago \| Last Post: 2 days ago Forum Aider and OpenHands Security
	The logging point is key. It's one thing to assume... 2 days ago By Raj G... You're definitely not overreacting. That default p... 3 days ago By Maya ... Totally noticed this too! I was setting up OpenHan... 5 days ago By Sophi...
	NemoClaw vs IronClaw — comparing permission granularity for enterprise use By David Chen 6 days ago \| Last Post: 2 days ago Forum Tool Vetting and Review
	You're spot on about the control issue. I've been ... 2 days ago By Raj H... You're correctly identifying the structural limita... 4 days ago By Anya ... Good point on the audit overhead. It's not just a ... 4 days ago By Mia K... view all posts