Grok Bot expands capabilities to search, read, and monitor X
Grok Bot announced that it can now search, read, and monitor X.
#ai-agents
View all AI newsNVIDIA researchers introduce PivotOPD to train AI agents on error recovery
NVIDIA researchers developed PivotOPD, a method designed to teach AI agents how to avoid early task mistakes and recover when errors occur. The training approach uses a teacher model to show agents better actions and guide them back on track over subsequent steps.
Study finds keeping agent harness prompts and tools small cuts costs by up to 3x
A study comparing Claude Code, mini-SWE-agent, and OpenCode with the same model on SWE-bench Verified found that minimizing harness prompts and tool schemas reduces costs by up to 3x without lowering accuracy.
CoreWeave introduces Agent Lens to group production agent failure traces
CoreWeave has detailed Agent Lens, a tool that analyzes an AI agent's full production trace corpus to cluster conversations with identical failure modes.
OpenCode uses agents to manage issues and pull requests
OpenCode is keeping its issues and pull requests active by deploying agents to respond to users, open pull requests for issues, and guide items to closure automatically.
Cline introduces Cline Cloud Agents to run coding tasks in the browser
Cline has launched Cline Cloud Agents, allowing users to run the AI assistant directly from a browser. Tasks run in a secure cloud sandbox to write code, test implementations, and open pull requests on GitHub repositories.
LangChain releases Managed Deep Agents v0.9 with Schedules SDK and per-run configuration
LangChain has released Managed Deep Agents v0.9, adding a Schedules SDK for autonomous task scheduling, per-run configuration across models and tools, and Slack reactions.
Michael Fan and andyisonline showcase new playground for GPT-Live
The tool is designed to help developers build intuition for the model's bidirectional interaction and its ability to delegate tasks to other models or agents.
Cloudflare Managed Defense uses specialized AI agents to analyze security alerts
Cloudflare Managed Defense uses AI agents built on Workers and network telemetry to analyze security alerts, separating deterministic evidence collection from model inference.
Microsoft Research introduces Agent Lightning to connect AI agents to RL training
Microsoft Research introduced Agent Lightning, a tool that connects existing AI agents to reinforcement learning training without requiring them to be rebuilt.
Viktor offers $100 credit trial for Slack agent evaluation assistant
Viktor provides an AI employee in Slack that analyzes AI agent harness evaluation logs, traces regressions to specific changes, and suggests rollbacks, offering a free trial with $100 in credits and no credit card required.
NVIDIA paper details VERA framework for co-evolving agent harnesses and models
An NVIDIA research paper presents VERA, a system that converts benchmark trajectories into over 9,000 restartable sandboxes with rubric scoring while co-evolving agent harnesses alongside model weights. The team open-sourced the environment corpus, reporting that a 27B agent scored 71.6 on AutoCoWorkBench to top Claude Opus 4.8.
Databricks details Omnigent policies for governing AI agents
Omnigent policies regulate agent cost and behavior by evaluating actions before execution, enforcing spending caps, restricting tool access, and requiring approval once accumulated session risk reaches a threshold.
Cursor enables remote control of desktop agents via iOS app
Cursor users can now monitor and control agents running on their computers directly through the Cursor iOS app.
Google and MIT introduce Coco agent platform for hardware and ML model co-design
Researchers from Google and MIT have introduced Coco, an agent platform designed for TPU architects to co-design machine learning models and accelerators using relational database simulation sweeps and typed tools.
CoreWeave launches Forge to connect agent tracing, evaluation, and serverless RL
CoreWeave has launched CoreWeave Forge, an environment that links agent tracing, evaluation, and serverless reinforcement learning. The platform offers a free starting tier and a 30-day Pro trial on new accounts.
WHOAFRO uses Google Earth AI agent prototype to map Ebola risk in DRC
WHOAFRO deployed Google Earth AI's Geospatial Reasoning agent prototype during the Ebola outbreak in the Democratic Republic of Congo, identifying 48 exposed settlements and over 45,500 at-risk people in minutes.
Every releases Slack agent built on Claude Managed Agents to subscribers
The Every team built a company agent on Claude Managed Agents for Slack to share skills across new models, making it available to subscribers following internal use.
Zichen Chen and Hugging Face open community voting for RSI Arena
Zichen Chen has launched the next stage of RSI Arena with Hugging Face, opening community voting on models trained autonomously by AI agents. Hugging Face Inference Endpoints serve each model live and stream evaluation feedback back to the agents for subsequent training.
Harness-Aware Distillation trains small model agents to 63.4% success on ALFWorld
A research paper presents Harness-Aware Distillation, a training framework that teaches small language model agents to leverage harness information by contrasting teacher decisions. The resulting student model reached 63.4% on unseen ALFWorld tasks, outperforming its 8B teacher.