The LGTM (Page 93)

The LGTM

Sign in Subscribe

Hermes v0.14.0 Turns Grok Subscriptions Into Agent Infrastructure

Hermes v0.14.0 Turns Grok Subscriptions Into Agent Infrastructure

Grok’s most important release this week may not be a model card, benchmark chart, or Elon post. It may be a plumbing release in an open-source agent runtime. Nous Research shipped Hermes Agent v0.14.0 on May 16, and the xAI headline is deceptively small: Hermes now supports

OpenAI Agents SDK Turns Absolute Sandbox Paths Into a Coding-Agent Trust Boundary

OpenAI Agents SDK Turns Absolute Sandbox Paths Into a Coding-Agent Trust Boundary

The OpenAI Agents SDK just patched the kind of bug nobody puts on a conference slide and everybody depends on in production: a sandbox path invariant. PR #3422 changes WorkspacePathPolicy so it rejects relative sandbox workspace roots. The stated problem is simple: the helper methods are documented and used as

DeepAgents Code 0.1.1 Fixes the URL Fetch Boundary Coding Agents Cannot Hand-Wave

DeepAgents Code 0.1.1 Fixes the URL Fetch Boundary Coding Agents Cannot Hand-Wave

The most important DeepAgents Code release this week is not the one that made the product line cleaner. It is the one that fixed a URL-fetching tool. LangChain released deepagents-code==0.1.1 with two patches: a corrected LangSmith sandbox working directory and a guard for the fetch_url tool

Google’s AI Education Labs Are an Enterprise Rollout Pattern in Disguise

Google’s AI Education Labs Are an Enterprise Rollout Pattern in Disguise

Google’s latest AI education post looks, at first glance, like another responsible-AI program announcement: workshops, stakeholders, roadmaps, the usual institutional vocabulary. That would make it easy to skip. It would also miss the useful part. Under the education framing, Google is describing the rollout pattern every serious organization now

NVIDIA’s Tiny Nemotron-CLIMB Models Are for People Who Cannot Afford to Guess at Scale

NVIDIA’s Tiny Nemotron-CLIMB Models Are for People Who Cannot Afford to Guess at Scale

NVIDIA’s Nemotron-CLIMB release is not trying to impress anyone in a chatbot arena. That is why it is worth paying attention to. The company published Nemotron-CLIMB Proxy Base Models on Hugging Face: two small decoder-only language models, one with 62 million parameters and one with 350 million, trained from

Codex-Spark Is OpenAI Splitting Coding Models Into Fast Hands and Slow Brains

Codex-Spark Is OpenAI Splitting Coding Models Into Fast Hands and Slow Brains

OpenAI’s most interesting coding-model launch this week is not another bigger brain for unattended pull requests. It is a smaller model designed to make the human stay in the loop because the loop finally feels fast enough. GPT-5.3-Codex-Spark is a research-preview variant of GPT-5.3-Codex aimed at real-time

GitHub’s Token Override Header Is a Small API Escape Hatch for a Very Real Agent-Runtime Dependency

GitHub’s Token Override Header Is a Small API Escape Hatch for a Very Real Agent-Runtime Dependency

The most dangerous assumptions in an agent runtime are rarely the cinematic ones. They are not “the model became sentient” or “the bot rewrote production.” They are usually a regex, a database column, a proxy, or a secret scanner that quietly assumed an access token would always look the same.

GPT-5.5 on Databricks Is a Reminder That Enterprise Agents Fail on PDFs Before They Fail on Philosophy

Enterprise agents do not usually fail first on philosophy. They fail on PDFs. They fail on scanned pages, legacy files, tables that lost their structure in 2014, invoices with one smudged digit, contracts with duplicated sections, permissions that hide the important appendix, and document sets where the answer depends on