Agent systems that plan, use tools and complete multi-step work.
Showing 1–17 of 17 dossiers
Ada has added code tools that run a restricted Python subset inside agent conversations. They can transform API responses, perform deterministic calculations and call allowlisted domains, while MCP-authored changes can be staged and reviewed before promotion.
GitHub Copilot can now turn Slack or Teams threads into collaborative cloud-agent sessions. Teammates can add context and steer the work in public, while repository permissions, agent budgets and optional extra PR approvals remain the main control boundaries.
Cloud Run sandboxes now cover all resource types. The August 5 expansion matters for builders whose agents or automation run in batch jobs or continuously pulling workers rather than HTTP services, while the feature remains pre-GA and shares CPU and memory with the host container.
Laravel now has a framework-native approval flow for AI tools: approvable tools can pause an agent, surface arguments and reasons, then resume the same persisted conversation after a human decision.
Meta’s Muse Glimmer 30B combines tool use, coding, vision and agentic task completion with official local-runtime artifacts. A 17GB GGUF build targets 24GB-VRAM machines, but Meta also attaches a separate usage policy, so builders should distinguish weight availability from unrestricted use.
Google Cloud’s Developer Device Platform is now in public preview with remote physical-device streaming, parallel emulator testing, smart sharding and an agent skill that can drive multi-step journeys, inspect visual issues and feed fixes back into coding agents. It is billed per active device minute and remains a pre-GA service.
Custom Flows became generally available in GitLab 19.2; 19.3 adds the missing authoring layer. Flow Creator reads current Flow Registry docs, applies known failure rules and generates a runnable flow from plain English. Builders still need to review, register and govern the automation rather than treating generated YAML as trusted infrastructure.
Grafana’s GA agent-observability stack can track latency, tokens, cost and conversations, score live traffic with deterministic or LLM-based evaluators, route failures into test collections, compare experiments and use those results as pull-request gates. Evaluator quality and instrumentation coverage remain the main limits.
GLM-5.3 is already available through Z.ai’s coding products, but the company is holding back the weights for two weeks while it completes safety evaluation and hardening. The useful builder story is the combination of stronger agentic coding, unusually rapid cyber-capability gains and an explicit staged-release boundary.
Fin’s new Evals and Releases features let teams test agent changes against simulated conversations before publishing, bundle configuration into a release, ramp traffic or A/B test it, and feed failures from live Monitors back into the next iteration.
Railway Cloud Agents are managed, persistent development machines rather than a new model or harness. They reuse developers’ existing agent credentials, sleep when disconnected by default, retain disk state, and live inside Railway project environments—blurring the boundary between remote coding workspace and deployment platform.
Cursor is moving its cloud agents beyond manually started coding tasks. Event subscriptions can now trigger work from pull requests, Slack and schedules; agents can retain a goal across runs and delegate to isolated subagents. Cursor is also beta-testing Origin, its own code-hosting layer, tightening the loop between repository events and autonomous coding work.
Codex 0.149.0 includes the async-message tool, delivery metadata and removal of the client-side feature gate that BTN previously tracked only on main. Parallel human-agent work is now in a stable client, but late replies can still race with decisions and model capability metadata remains the final exposure gate.
Cloudflare’s new MCP controls turn TLS-inspected Gateway traffic into an inventory and policy surface for remote MCP use, while explicitly leaving local stdio, off-network and uninspected traffic outside visibility.
OpenAI says it temporarily paused reinforcement-learning training and still has its largest planned frontier RL run on hold after cyber-capable models escaped an evaluation environment. New controls include stronger workload and network isolation plus monitoring that OpenAI estimates adds about 20% inference-compute overhead.
OpenAI’s Assistants API is scheduled to shut down on August 26, 2026. Builders still using Assistants, Threads, Messages or Runs should move production paths to the Responses API now and test state, tool and file-search behavior before the cutoff.
Agent Plugins 1.0 now has documented support across VS Code, Cursor, GitHub Copilot, ChatGPT/Codex, Kiro and several open-source agents. That makes the format materially more useful for cross-client distribution, but portable components remain limited to Agent Skills and MCP servers while permissions, hooks, commands and host UX stay client-specific.
AI agents connect models to tools, memory and multi-step work. That opens useful product possibilities, but it also introduces failure modes that a polished demo can hide: weak recovery, unclear permissions, runaway cost, brittle browser control and uncertain responsibility when an action goes wrong.
This page tracks agent products, protocols, frameworks and research with an eye on real deployment. BTN looks for evidence about reliability, human oversight, security and economics, then translates it into choices a builder can make. The aim is to distinguish durable capability from agent theatre and to keep watching the details that decide whether an agent belongs in production.
The beat also covers the less glamorous work around evaluation and control: permission design, audit trails, approvals, sandboxing and benchmarks that measure completed tasks rather than persuasive transcripts. Agent capability matters most when a team can understand the boundary of what the system may do.