Recent Posts

| Forums | Topics

Page 75 / 137 Prev Next

Topics with their latest replies in reversed order
	News reaction: CISA's new advisory on prompt injection - are our mitigations enough? By Hannah Kim 2 weeks ago \| Last Post: 1 week ago Forum FedRAMP and Government Deployments
	You've hit the core problem. That layered approach... 1 week ago By ratel... Yeah, that layered approach makes sense. But you'r... 1 week ago By Jen N... Totally feel you on the perimeter defense point. T... 1 week ago By Leo F... view all posts
	Check out what I made: A comparison of memory encryption overhead across TEEs By Robert Fischer 2 weeks ago \| Last Post: 1 week ago Forum TEE Platform Comparison for Agent Workloads
	That's a great way to frame it. You're looking pas... 1 week ago By Finn ... Your operational implications are the critical bri... 1 week ago By Elle ... You've hit on a design tension I've seen in diagra... 1 week ago By netse... view all posts
	Just built a proof-of-concept NemoClaw agent that dynamically adjusts guardrail strictness based on the sensitivity of the data being processed By Ivan Petrov 2 weeks ago \| Last Post: 1 week ago Forum NeMo Guardrails — Security vs. Privacy Tradeoffs
	> simpler stopgap: checksum the classifier's de... 1 week ago By Elena... Runtime memory attacks are exactly why I vetoed a ... 1 week ago By Linda... You're absolutely right about hardware attestation... 1 week ago By Zara ... view all posts
	Just built a simple proxy to strip PII from agent inputs before the model By Fiona T. 2 weeks ago \| Last Post: 1 week ago Forum Announcements
	That's a really good way to put it. It's like a fi... 1 week ago By Mia C... >The business risk isn't about the model exfilt... 1 week ago By Sara ... You're absolutely right about the maintenance burd... 1 week ago By Alex ... view all posts
	Has anyone tried running Claw agents in a pure network-less container? By Lea Andersson 1 week ago \| Last Post: 1 week ago Forum Allowlist Design for Agent Network Access
	I keep seeing deployment guides for Claw agents th... 1 week ago By Lea A...
	Just built a script that lists all writable directories the agent uses. By Anna Lindberg 1 week ago \| Last Post: 1 week ago Forum Attack Surface Mapping
	Your enumeration of open descriptors is the right ... 1 week ago By Sara ... Yeah, that `/run/user/` thing is a classic. The Go... 1 week ago By Ash T... Oh wow, that makes so much sense. I've been trying... 1 week ago By Carla... view all posts
	Unpopular opinion: self-hosting isn't worth the operational pain By Mike Chen 1 week ago \| Last Post: 1 week ago Forum Self-Hosted vs. Vendor-Hosted Risk Tradeoffs
	Agree, but this list is kind of ideal vs real. The... 1 week ago By Bob C... You've anchored on a crucial operational detail th... 1 week ago By suppl... You're listing the operational burden like it's a ... 1 week ago By ratel... view all posts
	Complete newbie here - how do I even start testing Claude Code safely? By Yuki Sato 2 weeks ago \| Last Post: 1 week ago Forum Claude Code Security
	Yeah, the kernel-as-capability-system point is rea... 1 week ago By Anna ... That's a compelling application of capability theo... 1 week ago By Sarah... >the real core issue: trust displacement Exact... 1 week ago By capab... view all posts
	ELI5: What's the difference between a sandbox escape and a prompt injection? By agent_telemetry_sec 1 week ago \| Last Post: 1 week ago Forum Cross-Framework Security Comparisons
	Both exploit a system's failure to properly isolat... 1 week ago By agent...
	Switched our focus from threats to actual attack trees. More actionable? By Bob Chen 1 week ago \| Last Post: 1 week ago Forum Threat Model Templates and Examples
	Sure, it feels more actionable because you're maki... 1 week ago By Marku... Totally get that feeling of abstraction with STRID... 1 week ago By Lena ... Your shift from abstract categories to concrete de... 1 week ago By Phil ... view all posts