LLM Digest

Story

arxiv_agent_systems_research · Nov 10, 2026 · paper

Source brief

From Attack Surface to Defense Architecture: Compositional Security for Agentic Coding Harnesses

arxiv.orgNov 10, 2026
original source linked

In brief

AI coding harnesses couple language models to shells, filesystems, and network sockets, turning the harness into an automated Living-off-the-Land (LotL) platform: once an adversary injects instructions into retrieved...

Feed lens

agenticharness

Read the original at arxiv.org →Open in live feed Read that day’s brief

From Attack Surface to Defense Architecture: Compositional Security for Agentic Coding Harnesses

Earlier in this thread 4 items

Using Claude to reduce the attack surface of code

Context as Infrastructure: Knowledge Compilation for Retrieval in Agentic Systems with Harnesses

Article: The Evolution of Agentic Harnesses: From Chains to Graph to Code

Can Agent Harnesses and Inference Engines Hear Each Other? The HEAR Protocol for Agentic LLM Serving