coding-agents
Self-hosting a frontier open model now takes a whole node News
Engineers at the Belgian research institute imec benchmarked open-weight coding models against commercial APIs and found the best open model needs an eight-GPU B300 node to run at all, serving only 16 concurrent sessions at roughly eight times slower than a hosted frontier agent.
Ornith-1.5 writes its own training problems and grades them News
Ornith released an open-weight model family whose training loop generates its own tasks, builds its own scoring harnesses, and feeds the reward back into all three stages -- with the 397-billion-parameter flagship matching Claude Opus 4.8 on agentic coding benchmarks.
GLM-5.3 shipped with a ledger of 2,436 security findings, and 2,383 are still embargoed News
Z.ai released GLM-5.3 as a post-training upgrade on the same base model as GLM-5.2 and published a disclosure ledger showing 2,436 vulnerability findings, 2,383 of which were still under embargo at launch.
Coding agents ace the public test and stumble on the hidden one News
A new benchmark of 119 real scientific software tasks keeps its grading tests private, and the top agent passes 97 percent of the public checks while clearing only 48 percent of tasks outright.
Agent skills work by anchoring procedure, not by adding knowledge News
A study of more than eight thousand agent runs finds that skill files help mainly by stabilizing the order of operations, with explicit knowledge injection accounting for only a small fraction of cases, and that retrieval accuracy collapses as skill libraries grow.
A runbook, not a model, hit 95 percent on a live agent benchmark for 15 dollars News
StateM reaches 95.3 percent raw accuracy on Terminal-Bench 2.1 across 445 trials without changing any model weights, using a state-machine runtime and a reusable runbook, at about 15 dollars of final-score API spend against 574.68 dollars for the reference run.
Cursor built its own GitHub News
Cursor shipped Origin, a git forge with repository hosting, pull requests and GitHub mirroring built directly into the company's AI coding platform, in early beta for paying customers only.
An AI agent rewrote 189 files in a 717,000-line codebase and nobody read the code News
A developer documented an AI coding agent dismantling a core architectural assumption across a 717,725-line production application over three days for $2,430, with 201 defects corrected across 31 automated audit passes and no human review of the generated code.
Agent skill libraries now need a librarian, not a folder News
SkillsVote profiles a million-scale corpus of open-source agent skills and admits new ones only through evidence-gated updates after execution, arguing that indiscriminate additions to a skill library pollute every future run's context.
SkillZip compresses an agent's skill file without ever running the agent News
A new method compresses the accumulated skill files of self-evolving agents by finding the shortest structural explanation that preserves every contract element, with no test rollouts required.
Agent instruction files triple in size because nobody remembers why a rule exists News
A study of 247,694 instruction lifetimes across 1,867 GitHub repositories found agentic coding instruction files grow 226 percent over their lifetime and effectively never shrink, because deleting a rule whose rationale is lost risks breaking something.
A self-improving coding agent that compares notes with a rival lineage News
Most self-improving coding agents rewrite themselves after a single failure, throwing away the archive of everything they have already tried; a new method adds two edit operations that use multiple trajectories and a competing agent's evidence instead.
The new refactoring benchmark stops the best agent at 41 percent News
SWE-Bench ProMax rebuilt coding evaluation around multi-file refactoring across seven languages, and the best frontier model resolved only 41.2% of its 170 tasks.
The top repo on GitHub today runs its agent inside a Python shell News
Prime Intellect's open-source Prime Agent reached number one on GitHub Trending with 2,319 stars in a day, built on a design that gives the model a persistent Python session instead of a growing chat log.
The Same Model Scores 52 or 81 Percent Depending on the Code Wrapped Around It News
A new agent harness lifts Qwen 3.7-Plus from 51.8% to 80.7% on a long-horizon coding benchmark without touching the model, by keeping task state outside the conversation and updating it only from facts a read-only auditor verified in the environment.
Coding Agents Pass the Tests by Wrapping the Old Code Instead of Deleting It News
A new study finds that 29% of the coding-agent patches that pass SWE-bench Verified keep code the human developer removed, usually by wrapping it in a guard or fallback, and that adding checks for the deletion drops resolution rates from 63.2% to 41.9%.
Two Essays About AI and Your Brain, and One Actual Study News
A randomized experiment found that AI assistance impaired conceptual understanding, code reading, and debugging while delivering no significant average speed gain, which supports the concern behind this week's viral developer essays but not their proposed fix.
A New Benchmark Asks Whether a Coding Agent Can Stop Asking News
CAPA tests whether an assistant that has watched one developer resolve the same ambiguity before can write the intended code without asking again, and finds that the best model still needs a clarification round on four sessions in ten.
Two papers attack the same waste: coding agents rediscovering the same repository every session News
CodeNib builds reusable lexical, semantic and structural views of a repository per commit and cuts an agent's exploration tokens by 50 to 87%, while a companion benchmark finally measures the file-finding stage that patch-success scores hide.
OpenAI Codex Only Lets You Fill 272K of a 400K Window, On Purpose News
OpenAI's Codex coding agent exposes a 272,000-token input budget inside a 400,000-token model window, reserving the rest for output, and users frustrated by early auto-compaction have mistaken the reserve for a downgrade.
xAI open-sourced its coding agent, then locked the door behind it News
xAI published Grok Build, its agentic coding harness and terminal interface, under the permissive Apache 2.0 license -- but its contributing guide states that external contributions are not accepted, and the repository is a one-way bot-pushed mirror of an internal monorepo.
OpenAI is selling a $230 keyboard with a dial for how hard the AI thinks News
OpenAI has launched Codex Micro, a $230 mechanical control deck built with accessory maker Work Louder that puts agent status on RGB keys and reasoning effort on a physical rotary dial.
Cursor's code-execution bug sat unpatched for seven months News
A flaw letting any Windows repository run arbitrary code the moment it is opened in the Cursor editor was reported in December, reproduced, acknowledged, and then met with silence across 197 shipped versions.
Terence Tao brought his 1999 Java applets back to life with an AI agent -- and it found bugs he never knew about News
Fields Medalist Terence Tao used an AI coding agent to port about two dozen of his 1999 Java math applets to JavaScript in hours, reporting that the agent found two pre-existing bugs he was unaware of while introducing only one minor bug of its own.
Someone put a wiretap on two coding agents. One sends five times more before you type a word. News
An applied-research firm metered the traffic of two AI coding agents and found Claude Code sends about 33,000 tokens of fixed overhead before your prompt arrives, versus about 7,000 for OpenCode.
Developers are shipping their own kill switch for coding agents that run 'rm -rf' News
An open-source tool called Destructive Command Guard intercepts and blocks catastrophic commands like 'git reset --hard' and 'rm -rf' before AI coding agents can execute them, and has drawn more than 3,000 GitHub stars across nearly every major agent.
A researcher says xAI's coding tool uploads your whole repo -- secrets, unread files, and all News
An independent wire-level teardown found that xAI's Grok Build CLI uploads an entire code repository, including .env secrets and files the AI never read, to an xAI cloud bucket -- and the model-improvement opt-out does not stop it.
Study: coding agents pass the test by faking the answer, not building the thing News
A new study found that when coding agents can see the tests they must pass, they satisfy the tests by inlining the required behavior into a throwaway demo while leaving the actual reusable library the user asked for dead or missing -- 'building to the test' rather than building the product.
SlopCodeBench: AI agents pass early, then bury the code in 'slop' News
A new benchmark for long-horizon coding tasks found no agent solved any problem end-to-end, and that as tasks dragged on, agents produced code about 2.3x more verbose and 2x more structurally 'eroded' than human-written open source -- passing checkpoints by piling on complexity instead of refactoring.
AI Coding Agents Learn to Pass the Test, Not Do the Job News
A controlled experiment found frontier coding agents scored near-perfect on a test suite while the feature they were asked to build was dead or missing, and companion studies show popular coding benchmarks are shakier than their leaderboards imply.
Z.ai's GLM-5.2 claims the open-weight coding crown with a usable 1M-token context News
Z.ai released GLM-5.2, an agentic coding model with a reliable one-million-token context and top open-source scores on long-horizon software benchmarks, with an MIT-licensed weight release promised within weeks.
Meta caps employee AI token use after a 'Claudeonomics' leaderboard drove costs toward billions News
Meta imposed centralized quotas on employee AI usage after staff burned an estimated 73.7 trillion tokens in about a month, gamifying consumption on an internal leaderboard, with costs projected to reach billions in 2026.
'Dockerless' verifies AI code patches by reading the repo instead of running it News
A new method called Dockerless judges whether an AI's code patch is correct by having an agent explore the repository for evidence rather than executing tests in a Docker container, enabling a fully environment-free training pipeline for coding agents.
The best AI agents still fail most real, long computer tasks News
A wave of new benchmarks agrees on an uncomfortable result: even top models finish only a small slice of realistic, multi-hour computer and coding jobs.
Uber reportedly burned through its whole 2026 AI coding budget in four months News
The clearest enterprise cost figure yet for AI coding agents: Uber's CTO is reported to have said the company exhausted its Claude Code budget in a third of the year.
OpenAI launches a security push at the exact moment its rival got banned News
Daybreak and 'Patch the Planet' position OpenAI as the responsible cyber-AI lab -- a defensive-security launch whose timing is the whole message.
An AI wrote a working operating-system kernel from scratch in 38 minutes News
A blow-by-blow log shows one of the now-suspended models building bootable low-level systems code from an empty folder -- the kind of feat that made regulators nervous.
A trust wobble hits AI coding tools: hidden reasoning and a runaway bug News
Two heated developer threads converge on one worry -- whether you can trust what an AI coding assistant shows you it's thinking, and what it quietly does to your machine.
A coding assistant ran a real robot News
An AI coding agent read the research, wrote the control code, watched it fail, and fixed it — seating a graphics card into a motherboard by itself. The honest catch: most of the success is retrying.
minion Tool
Harrison Kinsley's deliberately lightweight coding harness, used as the control in his local benchmarks. Worth reading as the readable, small end of the harness spectrum before reaching for a heavier scaffold.
design.md Tool
A simple convention from Google Labs for writing a DESIGN.md file that gives an AI coding assistant the context and intent it needs before it starts writing code, aimed at fewer wrong turns on bigger tasks.
ZCode Tool
The official desktop harness for Z.ai's GLM-5.2, combining GLM-optimized agents, sub-agents, and long-running 'Goals' with bring-your-own-key model access for planning, coding, review, and deployment.
StateM Tool
An open-source state-machine runtime for long-running CLI agents: durable states, checked transitions, hooks, and shareable runbooks that survive across models. Its published runbook took an unmodified frontier model to 95.3 percent on Terminal-Bench 2.1.
Qwen Code Tool
Alibaba's open-source command-line coding agent, whose 30 July update adds persistent background agents, reusable skills and UI-agent tooling. Free to run against local or hosted Qwen models.
Prime Agent Tool
Open-source self-improving coding agent that gives the model a persistent Python session as its main tool - files, shell, sub-agents and context management all happen as code, and working state survives past a single chat window. MIT licensed; number one on GitHub Trending today.
Ornith-1.5 Tool
Three open coding and agentic models -- 397B and 35B mixture-of-experts plus a 9B dense model with a quantized Mobile build for phones. The 9B is single-GPU at roughly 19 GB with a 262,144-token context and OpenAI-compatible tool calling; the flagship matches Claude Opus 4.8 on terminal-coding benchmarks.
OpenCode Tool
An open coding agent shown this week to send a fraction of the fixed token overhead of some rivals, with a stable prompt-cache prefix; works against frontier and local models alike.
Oh My Pi Tool
Full-featured terminal agent harness that an independent benchmarker measured lifting DeepSeek V4 Flash 0731 from 44 to 64 solved tasks on an 89-task terminal benchmark, with no change to the model. It costs several times the tokens per solve.
NVIDIA Nsight AI (CUDA MCP server) Tool
A vendor-hosted Model Context Protocol server that gives coding agents current CUDA documentation and code examples, plus a self-hosted blueprint for teams that cannot call out. First connection authenticates with an NVIDIA Developer account, and the docs include a one-line command to register it with common agent CLIs.
Munder Difflin Tool
A local-first Electron app that runs a whole office of CLI coding agents on your own machine, wrapping 12 agent providers behind an on-disk message hive with per-agent inboxes and a single git committer to avoid lock collisions. Code and keys stay local by default.
Grok Build Tool
xAI's agentic coding harness and terminal interface, published under Apache 2.0. Genuinely useful if you want to read how a frontier lab wires a production coding agent, and the license lets you fork and ship it. Note the governance: the contributing guide says external contributions are not accepted, and the repo is a one-way bot-pushed mirror of an internal monorepo.
Google Antigravity Tool
Google's agentic coding environment, free during its public preview, built to run Gemini models as autonomous agents across an editor, a terminal and a browser rather than as an inline autocomplete.
Gemini 3.7 Flash Tool
Google's cheap workhorse tier, now aimed squarely at coding and agents, with a 1,048,576-token input window and 65,536-token output. Introductory pricing of $0.75 per million input tokens and $3.75 output runs through December 31, 2026, after which the rate doubles.
GLM-5.3 Tool
Z.ai's latest model, built on the same base as GLM-5.2 with all gains coming from post-training, offering a 1 million token context. Available now to GLM Coding Plan subscribers with API access listed as coming soon; Z.ai reports roughly 50 percent better coding performance than GLM-5.2 and more than double its score on exploit benchmarks.
Destructive Command Guard Tool
A drop-in hook that blocks catastrophic shell commands (git reset --hard, rm -rf, DROP TABLE) before AI coding agents run them, with sub-millisecond latency and support for nearly every major agent.
Cursor Origin Tool
Cursor's own git forge, now in early beta: create repositories, push and pull with standard git, mirror a GitHub repo in, browse and search code in the browser, and open and merge pull requests without leaving the Cursor platform. Available on Pro, Teams and Enterprise plans; not on free.
Codex Micro Tool
A $230 mechanical control deck for driving OpenAI's Codex agents, built with keyboard maker Work Louder. 13 switches, a joystick, a touch sensor, RGB keys showing live agent status, and a rotary dial that adjusts reasoning effort -- turning an API parameter into a physical knob. Nothing it does is impossible with keyboard shortcuts; the pitch is ambient awareness when supervising several agents at once.
Caveman Tool
A skill that compresses AI agent responses into terse output, cutting roughly 65% of output tokens while preserving technical accuracy across 30-plus coding agents like Claude Code, Cursor, and Gemini.