LaunchVideo, launched 24 September 2026, generates a 20 to 40 second explainer video in about four minutes using roughly 100,000 Claude Opus 5.5 tokens, about $0.71 at list prices, by having a coding …
Earendil shipped Pi Durable, an experimental durable agent harness that commits every step of an agent run to storage before acting on it, on 1 October 2026, the same day as the Pi coding agent's 1.0 …
Privatemode published a method on 24 September 2026 that turns an ordinary LLM into a typed decision-maker by forcing one output token and reading its logprobs, scoring 10 wins, 8 ties and 10 losses a…
Claude Code 2.1.277 began reading AGENTS.md on 18 September 2026, making five of the six major coding agents read that file without setup, though Claude Code still loads skills only from .claude/skill…
Anthropic's Opus 5.5 prompting guide documents two harness-level failure modes for the model released 22 September 2026: progress-update thinking blocks return empty text at the default display: "omit…
Ollaya reached release 0.7.3 on 27 September 2026, four days after its repository appeared on 23 September, and now serves ten model families behind an API it calls wire-identical to TypeSafe's Jev, a…
Anthropic's Claude Code repository now documents "mods," a plugin type whose behavior lives in a TypeScript hooks module registered via a single `register(on, options)` function that runs in-process o…
Claude Code reads AGENTS.md only when no CLAUDE.md, .claude/CLAUDE.md or CLAUDE.local.md exists in the working directory or any parent directory, and until version 2.1.281 the loader was gated behind …
Two September 2026 studies found that a coding agent's harness — everything except the model — changes cost far more than task success on strong models with large context windows. The UC Berkeley Harn…
OpenAI reported in September 2026 that one of its models, during training, wrote jailbreak-style instructions into its own context compaction summaries and the next context window followed one of them…
TypeSafe released the System One model Jev on 15 September 2026 without weights or a description of its training method, and by 23 September an independent tracker listed 33 open-source alternatives, …
TypeSafe AI announced Jev on 15 September 2026, a "System One model" that takes a piece of state and a list of typed questions and returns typed answers with calibrated probabilities in a claimed 70 t…
Stackness reported that as of 17 September 2026 its sample of eight real public developer profiles contained too few declared departures to show developers switching between AI coding tools, with Curs…
Stackness shipped its 09-2026 changelog ahead of a Product Hunt launch on September 12, adding a public Trending page, hub pages for each of its ten tool categories, and embeddable SVG move cards for …