LLM Vulnerabilities

Feeds to Scour
SubscribedAll
Scoured 228 posts in 24.6 ms

Assessing Automated Prompt Injection Attacks in Agentic Environments

 💉Prompt Injection  Content type: Academic
arxiv.org·

Indirect Prompt Injection remains a fundamental security challenge for AI

 💉Prompt Injection  Content type: Blog
brave.com·

ChatGPT Introduces Lockdown Mode to Everyone, Preventing Prompt Injection Attacks

 🛡️AI Security
researchsnipers.com·

Mathematical proof reveals why fixed AI guardrails can never block every jailbreak

 💉Prompt Injection
techxplore.com·

ashp15205/guardian-runtime: A zero-latency, local-first runtime firewall for LLMs. Intercept every prompt and response locally to stop data leaks and runaway token costs.

 🤝Multi-Agent Orchestration  Content type: Code
github.com··Hacker News

Measuring Embedding Drift: Why Hybrid Search Saves Stale Models.

 🔗Hybrid Search
pub.towardsai.net
·

Tiberius: A Security Testing Framework for LLM Applications in Java

 💉Prompt Injection
foojay.io·

Prompt Injection in RAG Agentic Systems

 💉Prompt Injection
ulad.net··Hacker News

iOS 27 system prompts

 🔧Developer tools

From prompt to pwned: chaining LLM and web bugs to Admin

 🛡️AI Security  Content type: Blog
blog.quarkslab.com·

Humans and LLMs share a mental disorder: Fugue Lock

 🦉Qwen
vwwwv.org··Hacker News

OpenAI unveils Lockdown Mode to protect sensitive data from prompt injection attacks

 💉Prompt Injection

LLM Observability: What To Instrument and How To Act on It

 🪄Prompt Engineering  Content type: Blog
blog.n8n.io·

vishal-dehurdle/state-harness: Runtime safety net for LLM agents. Detects token spirals, kills doomed tasks early, tells you exactly why. Rust core, Python SDK. pip install state-harness

 🎭Claude  Content type: Code
github.com··Hacker News

Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation

 💉Prompt Injection  Content type: Academic
arxiv.org·

OpenAI Help: Lockdown Mode

 💉Prompt Injection
simonwillison.net·

Comparing Claude Fable 5's system prompt to Opus 4.8

 💻Claude Code  Content type: Blog

ChatGPT easily bypasses its own guardrails; all LLMs are inherently unsafe

 🎭Claude  Content type: Blog
techzine.eu·
Less-relevant results

Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭 (@elder_plinius)

 🎭Claude
xcancel.com··Hacker News

Purpose-built local AI agents

 🤖AI  Content type: Blog

Keyboard Shortcuts

Navigation

Next / previous item
j/k
Open post
oorEnter
Preview post
v

Post Actions

Love post
a
Like post
l
Dislike post
d
Undo reaction
u
Save / unsave
s

Recommendations

Add interest / feed
Enter
Not interested
x

Go to

Home
gh
Interests
gi
Feeds
gf
Likes
gl
History
gy
Changelog
gc
Settings
gs
Browse
gb
Search
/

General

Show this help
?
Submit feedback
!
Close modal / unfocus
Esc

Press ? anytime to show this help