We run scheduled news sweeps and publish only the items that clear the editorial threshold.
Message threads is in public beta on the Claude Platform. Today a coding agent resends every earlier message and tool result with each request.
The authors had the Kiro CLI agent, running Claude Opus 5.5 on Amazon Bedrock, write Lean 4 models of 26 functions across three codebases: the Kiro agent harness, the CLI client and Kiro Crew. The agent then wrote proofs and differential…
Adds onFailure: "block" for Agent SDK hook callbacks, so a callback that fails blocks the action instead of letting it run. Also adds a prompt before a web or shell tool call sends your account email address to the network, and a cost…
The authors built a dataset of 148 PyTorch research projects and had Claude Code, running Claude Opus 5.5, migrate ten of them to Google Cloud TPUs. They then checked each migration themselves instead of relying on the agent's report.
OpenAI says that around the end of last year its monorepo systems began struggling, with slow fetches and pushes, failed merges and reliability at times below 90%. Its explanation is that Codex lets each engineer have several changes…
An Agent Skills skill plus a 30-case benchmark of data-analysis traps. The author's own runs, one per case and arm and judged by Sonnet, show Claude Haiku 5.5 passing 37 of 39 graders on the 19 data-analysis cases without the skill and 39…
Factory made programmatic tool calling the default in Droid 1.0. The model writes JavaScript that runs in a sandbox with no file, network or system access of its own, and one model request can trigger hundreds of tool calls.
The authors reviewed sources published from February 2025 to September 2026: 58 practitioner articles, 11 company or vendor articles, five arXiv preprints, one industry report and one technical document. No forum threads passed their…
Momentic built the benchmark, ran it and graded it, and its own testing agent, Mo, comes out on top. The suite has 20 product areas in four open-source apps (Gitea, Metabase, Twenty and Saleor) with 1,396 labeled bugs.
CloudOpsBench v1.0 has 45 AWS tasks, including decommissioning projects, securing access, migrating services and repairing failures. The post says Claude Fable 5.1 completed 29 of the 45 and GPT-6 Astra completed 26.
The author says his company opened 694 pull requests in the last week of October, against 42 to 75 a week in June. Most were written by coding agents and opened under staff names.
The README describes a tool that sits between an agent and a database. Rules allow, deny or require approval for each tool call, and Deedglass then compares the agent's report with a read-back from the target.
The author ran the same coding agent (Claude Code) on one bug in the RISC-V core inside Caliptra, a hardware root of trust, ten times in each of two setups. The bug was a historical one in the VeeR EL2 processor, reintroduced by reverting…
SineFrame, which builds the open-source MCP testing tool M3, reports that Claude Haiku 5.5 answered 82.0% of tasks correctly from the server through OpenCode and 65.4% through Claude Code. The runs covered 21 MCP servers, about 20 tasks…
404 Media reports that in late August an employee in Proton's security Slack channel asked for a few days to pentest newly written code before it shipped. CEO Andy Yen replied: "Of course it's possible to be fast and secure." He called…
Lawson had a coding agent build one Pokémon app, using the PokéAPI, in twelve frameworks and meta-frameworks. He ran Opus 5.5 at high effort through Claude Code in a sandbox, five times per framework, and reports medians against a…
OpenAI's help page says that starting in December, the Plus, Pro and Business plans' usage allowance "will be shared across ChatGPT and Codex." Usage will be measured in tokens, and how much of the allowance a request uses depends on the…
The repository rebuilds the harness from GVS5H (Lee, 2026) as a Memdoor workflow: a planner, two ideators, a solver, a manager and a repair worker, each starting from a fresh context and passing work through files on disk. The README states…
In ViBench 2.0 the agent starts from an empty project, builds a first version of a product from a short spec, then adds up to eighteen more features on top of its own code. The benchmark has seventeen products, 241 feature requests and…
The release notes state that fast mode for Opus 4.8 is deprecated as of October 30, 2026 and will be removed on November 30, 2026. After removal, fast mode requests on Opus 4.8 return an error and do not fall back to standard Opus 4.8.