Skip to content
Navigation menu
Search
Powered by Algolia
Search
Log in
Create account
DEV Community
Close
#
llm
Follow
Hide
Posts
Left menu
đź‘‹
Sign in
for the ability to sort posts by
relevant
,
latest
, or
top
.
Right menu
Tool Grants Are Firewall Rules: How Excessive Agency Turns Permitted Actions Into Production Incidents
Davi
Davi
Davi
Follow
Sep 6
Tool Grants Are Firewall Rules: How Excessive Agency Turns Permitted Actions Into Production Incidents
#
agents
#
ai
#
llm
#
security
Comments
Add Comment
4 min read
AI Agents Are eval(userInput) With Extra Steps
Davi
Davi
Davi
Follow
Sep 6
AI Agents Are eval(userInput) With Extra Steps
#
agents
#
ai
#
llm
#
security
Comments
Add Comment
5 min read
Open-Source LLMs Are Unverified Dependencies: The Model Backdoor Gap npm Already Solved
Davi
Davi
Davi
Follow
Sep 6
Open-Source LLMs Are Unverified Dependencies: The Model Backdoor Gap npm Already Solved
#
cybersecurity
#
llm
#
opensource
#
security
Comments
Add Comment
5 min read
RAG Poisoning Is a Supply Chain Attack, Not a Prompt Injection Variant
Davi
Davi
Davi
Follow
Sep 6
RAG Poisoning Is a Supply Chain Attack, Not a Prompt Injection Variant
#
ai
#
llm
#
rag
#
security
Comments
Add Comment
4 min read
Loops, Cron, and Idempotency: Agents That Operate While You Sleep
Davi
Davi
Davi
Follow
Sep 6
Loops, Cron, and Idempotency: Agents That Operate While You Sleep
#
agentic
#
llm
#
loops
#
cron
Comments
Add Comment
4 min read
LLM Covert Channels: How AI Agents Exfiltrate Data in Plain Sight
Davi
Davi
Davi
Follow
Sep 6
LLM Covert Channels: How AI Agents Exfiltrate Data in Plain Sight
#
ai
#
cybersecurity
#
llm
#
security
Comments
Add Comment
5 min read
Prompt Caching Is a Timing Oracle: How the 41-80% Cost Win Becomes Cross-Tenant System Prompt Extraction
Davi
Davi
Davi
Follow
Sep 6
Prompt Caching Is a Timing Oracle: How the 41-80% Cost Win Becomes Cross-Tenant System Prompt Extraction
#
api
#
cybersecurity
#
llm
#
security
Comments
Add Comment
4 min read
MCP Tool Schema Poisoning: When the Framework Layer Authorizes the Attack
Davi
Davi
Davi
Follow
Sep 6
MCP Tool Schema Poisoning: When the Framework Layer Authorizes the Attack
#
cybersecurity
#
llm
#
mcp
#
security
Comments
Add Comment
6 min read
OPSEC for Agents: Scope, Caps, Monitoring, and Kill Switches
Davi
Davi
Davi
Follow
Sep 6
OPSEC for Agents: Scope, Caps, Monitoring, and Kill Switches
#
agentic
#
llm
#
opsec
#
security
Comments
Add Comment
4 min read
99.7% Rejected in 84ms: Why I Stopped Making the Generator Smarter
hidetzu
hidetzu
hidetzu
Follow
Sep 6
99.7% Rejected in 84ms: Why I Stopped Making the Generator Smarter
#
llm
#
verification
#
architecture
#
go
Comments
Add Comment
7 min read
The Real Cost of LLM APIs in 2026: A Developer's Field Guide
ninebox
ninebox
ninebox
Follow
Sep 6
The Real Cost of LLM APIs in 2026: A Developer's Field Guide
#
ai
#
llm
#
openai
#
webdev
Comments
1
 comment
3 min read
Regression-testing an agent whose output is never the same twice
Cuong Duong
Cuong Duong
Cuong Duong
Follow
Sep 6
Regression-testing an agent whose output is never the same twice
#
testing
#
ai
#
llm
Comments
Add Comment
3 min read
It Fit in Memory and Was Still Unusable — Do the Bandwidth Arithmetic First
John
John
John
Follow
Sep 6
It Fit in Memory and Was Still Unusable — Do the Bandwidth Arithmetic First
#
hardware
#
llm
#
machinelearning
#
performance
1
 reaction
Comments
Add Comment
5 min read
The Week AI Shipped With Locks On: GPT-6 Astra and Claude Fable 5.1
AgustaON
AgustaON
AgustaON
Follow
Sep 6
The Week AI Shipped With Locks On: GPT-6 Astra and Claude Fable 5.1
#
news
#
ai
#
openai
#
llm
Comments
Add Comment
7 min read
Agent Cost Optimization: Token Accounting, Cache, and Dynamic Dispatching
Davi
Davi
Davi
Follow
Sep 6
Agent Cost Optimization: Token Accounting, Cache, and Dynamic Dispatching
#
agentic
#
llm
#
cost
#
optimization
Comments
Add Comment
4 min read
đź‘‹
Sign in
for the ability to sort posts by
relevant
,
latest
, or
top
.
We're a place where coders share, stay up-to-date and grow their careers.
Log in
Create account