A study by Cassis on LLM context optimization for data analytics found that adding a full example library raised Haiku's accuracy from 70% to 82% and Sonnet's from 81% to 93%, while removing rules and…
Anthropic's Claude Code v2.1.139 (May 2026) introduces the /goal command, which lets users set a completion condition so the AI autonomously runs multiple turns until an independent evaluator confirms…
Anthropic's prompt caching, Batches API, and model routing can significantly reduce LLM costs without switching providers. Prompt caching reuses prefixes at 0.1× the base input price, the Batches API …
BoundFlow, an open-source control plane for AI agents, enforces cost caps, approval gates, and self-healing policies on unattended LLM workflows. The backend is Apache-2.0 licensed and self-hostable, …
Boris Cherny, creator of Claude Code at Anthropic, introduced /goal and /loop commands that enable autonomous AI agents to work without human intervention, shifting the paradigm from prompting to loop…
Databricks released an internal benchmark evaluating coding agents on its multi-million line codebase, finding that models like GLM 5.2 offer top-tier performance at lower cost than alternatives like …
A developer benchmarked 14 AI agent models and found that the true cost of a simple greeting like "Hi" is not the token price (as low as $0.005) but the waiting time, which can cost up to $5.10 per gr…
Developer Kenimo49 released compact-ops, a Claude Code plugin that protects AI coding sessions from context compaction by preserving operational state. During the release, the plugin successfully warn…
SerpApi developed an agent usability testing (AUT) skill that uses a second AI agent to observe and score how well LLM agents follow agentic skills, achieving 100% task success with the skill versus 2…
Vercel's open-source eve agentic framework allows developers to turn a folder of agent files into a deployed app, handling durable runs, sandboxed compute, approvals, and tracing. The author, who runs…
Fable, a senior decision-making AI agent, is designed to preserve premium reasoning for high-value tasks such as intent understanding, architecture, decomposition, tradeoffs, final review, and synthes…
A new open-source stack of 98 patentable AI architectures packaged as Claude Code skills achieves 93% of Fable 5 quality at 1/125th the cost by using Haiku models with optimized prompting and architec…
Anthropic's Claude model lineup—Haiku, Sonnet, and Opus—offers different trade-offs between speed, cost, and capability. The article provides a straightforward rule for choosing among them based on ta…
An engineer at a company running headless Claude Code instances on Linux discovered that a static OAuth access token can return HTTP 200 on a raw API call while a long-running instance using the same …
Researchers developed a method to combine AI control protocols by routing between them based on predicted usefulness and suspicion, achieving better safety-usefulness tradeoffs. They also addressed at…
An engineer at Anthropic analyzed the hidden costs of switching models in Claude Code sessions, finding that model switches invalidate prompt caches, leading to duplicated cache writes that can increa…
A developer reports that AI coding costs can be slashed by 70% through task-level model routing instead of using a single frontier model for all tasks. After profiling usage, the developer found that …
A developer building an ad analytics SaaS reduced Claude API costs from $180-200/month to $95-110/month by implementing a three-tier routing system that assigns tasks to Haiku, Sonnet, or Opus based o…
A developer discovered that AI model performance degrades as context grows, with replies slowing and rambling after context exceeds 80%. The developer implemented two mid-session strategies: model tie…
A new open-source Git diff filter uses Claude AI to annotate every changed line with a one-sentence explanation inline in the terminal, helping developers understand code changes during interactive st…