Every processed story in chronological order, with the newest coverage first. Filter by tag, source, or score to drill in.
Both Claude Code and Codex have been writing granular token and cache-usage data to local disk all along, meaning developers can diagnose and fix prompt cache inefficiencies — the primary driver of hitting subscription limits — without any API call or provider dashboard.
This release resolves several MCP integration failures — including schema validation errors and timeout drops on long-running tools — that previously blocked OpenAI-compatible and Cloudflare AI Gateway users from reliably running MCP-powered workflows.
cwcode's hash-anchored edit scheme and sticky prefix-cache design directly cut token costs and output volume compared to naive agent loops, making sustained multi-hour autonomous coding runs on local or low-cost LLM endpoints practical without any cloud service dependency.
The stricter input validation closes a silent failure mode where malformed arguments were treated as prompts, and the new `cline skill` command brings skill management to parity with the existing plugin and MCP command surface.
The research introduces a structured framework for measuring Claude Code's real-world usage and task outcomes, providing a basis for tracking how the tool's impact evolves as adoption grows.
The agent merge feature removes the manual loop of copying code review feedback and CI failures back into prompts, letting the app resolve them autonomously on a monitored pull request.
By storing the complete agent state history inside the Git commit graph, mlx-code makes agent sessions inspectable and resumable with standard Git tooling instead of a proprietary database.
The update adds diff visibility and staged feedback directly into Amp's agentic coding threads, addressing the human review step that @beyang identifies as the current bottleneck.
GameCraft-Bench exposes a concrete ceiling on current coding agents' ability to produce fully playable games, showing that even the best frontier models fall below 41.46% on a task requiring integrated scripts, scenes, assets, and runtime interaction — a gap that partial code-generation benchmarks do not capture.
The gateway directly addresses the gap between giving engineering teams free choice of coding agents and maintaining organizational visibility and control over the resulting LLM spend.