Claude Sonnet 4.6

Named by 6 kept items in this workspace.

The items that name it 6

Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox
writeup2026-10-22
Same Simulation, Different Evidence: Where Retrieved Tool Output Costs LLM Agents Accuracy
paper2026-10-19
SpatialChain: A Benchmark for Auditing Spatial Reasoning Faithfulness in VLMs
paper2026-10-05
NEUROTESTGEN: Neuro-Symbolic Guided Test Generation with Large Language Models
paper2026-09-24
Harness Engineering for Self-Improvement
article2026-07-04
rikugan: LLM reverse-engineering agent for IDA Pro and Binary Ninja
repo

In every workspace you can read 19

AI

Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox
writeup2026-10-22
Same Simulation, Different Evidence: Where Retrieved Tool Output Costs LLM Agents Accuracy
paper2026-10-19
SpatialChain: A Benchmark for Auditing Spatial Reasoning Faithfulness in VLMs
paper2026-10-05
NEUROTESTGEN: Neuro-Symbolic Guided Test Generation with Large Language Models
paper2026-09-24
Harness Engineering for Self-Improvement
article2026-07-04
rikugan: LLM reverse-engineering agent for IDA Pro and Binary Ninja
repo

Security

Combining Long-Term Memory and LLMs Comes at a High Cost: From Prompt Injection to Persistent Compromise
paper2026-10-20It is written here as Claude Sonnet 4.6 (shared-memory features allow compromise of multiple users in an organization and exfiltration of user prompts).
GhostSplice: Malicious MCP Servers Split Exfiltration to Bypass Guardrails
article2026-08-19
Benchmarking Secure-and-Functional Remediation and How Snyk Agent Fix Lifts Frontier-Model Fix Rates by over 14%
article2026-08-18
Socket: New Study Identifies 53 Slopsquatting Targets Across 5 Frontier LLMs
article2026-07-23
AI Security Briefing
article2026-07-18
AI Security Briefing
writeup2026-07-18
AI Security Briefing
article2026-07-18
Snyk VulnBench JS 1.0: Can LLMs Find the Same Bugs Twice?
writeup2026-06-29
Endor Labs: AI Coding Agents Cheat Security Benchmarks via Training Recall, Not Reasoning
article2026-06-11
The sorry state of skill distribution
writeup2026-06-03
Claude Code Auto Mode
article2026-05-13
IronCurtain: Secure AI Agent Runtime
writeup2026-05-07
rikugan: LLM reverse-engineering agent for IDA Pro and Binary Ninja
repo

All technologies in this workspace.