How to Audit and Cut Your Coding Agent Token Costs with LangSmith
Coding agent bills spiking? LangSmith's 4-stage framework — visibility, standardization, optimization, governance — cuts costs fast.
How-Tos
525 piecesHow to Use LangSmith Traces to Systematically Improve Your AI Agents
LangChain's Vivek Trivedy: improving AI agents means treating trace data as a data mining problem — here's the framework.
OpenAI GPT-Realtime-2.1: The Voice Agent Migration Guide — Latency Cuts, Reasoning, and the Mini Model
OpenAI's GPT-Realtime-2.1 lands with 25%+ latency cuts, better tool use, and a mini variant — the next step for production voice agents.
SNU DualView: How Seoul National University Built a 100% Effective Indirect Prompt Injection Defense for OpenClaw Agents
SNU's DualView blocks 100% of indirect prompt injection attacks in OpenClaw agents via dual environment views — and ships as a ready-to-install plugin.
Claude Code v2.1.202 Released — /dataviz Skill, Dynamic Workflow Sizing, and Telemetry Upgrades
Claude Code v2.1.202 ships dynamic workflow agent-count controls and richer OpenTelemetry pipeline telemetry — plus a clarification on /dataviz timing.
Illinois Becomes First US State to Require Annual Independent AI Safety Audits — SB 315 Signed July 6
Illinois becomes the first US state to mandate annual independent AI safety audits for frontier developers under SB 315, effective January 2027.
MakerChecker: Open-Source AI Agent Security Scanner Launches with CI/CD Governance
MakerChecker launches as an open-source scanner for dangerous AI agent capabilities — CI/CD-ready with tamper-evident audit logs and runtime governance.
MCP 2026-07-28 Release Candidate: Sessions Dropped, Every Request Now Stateless
MCP RC drops sessions entirely — stateless HTTP core means every request is now self-contained; breaking change ships final July 28, 2026.
Tencent Hunyuan Hy3 Launches: 295B MoE Open Weights Model Built for Agentic Workloads
Tencent's Hy3 — 295B MoE, Apache 2.0, 74.4% SWE-bench — launches as a serious open-weights rival for production agent workloads.
How to Reconfigure Claude Code After the v2.1.200 Permission Mode Change
Claude Code v2.1.200 changed the default permission mode to Manual and disabled AskUserQuestion auto-continue. Here's how to update your headless workflows.