Models Outrun Warnings
A practical guide for reviewing AI models.
AI
Artificial intelligence, local AI, agents, tooling and AI security.
Agent Attacks Survive Session Resets
An operational review of agent-assisted work.
Verify the Bytes Your Agent Will Run
Executable trust needs runtime evidence.
Audit Who Writes Your Agent’s Rules
Before approving a dependency, ask who supplied the instruction and who verified the package. Our guide separates those decisions.
Before Approving Dental AI
Read the evidence, then decide what still needs testing. Our worked example follows the claim, the review and the acceptance decision.
The Warning Followed the Action
Inspect permissions and evidence before trusting a warning.
Patch the Gateway Your Feature Actually Uses
Verify each route and running build.
Test the Reachable Service
Keep evidence before closing tickets.
Access Does Not Approve a Patch
Make review accountable.
Two LightLLM Helper Ports Expose Code Execution on Inference Nodes
Two LightLLM helper services can accept unsafe network input outside the model API. Check multimodal and profiling nodes, their live ports and their network boundaries.