Simon Willison released llm-coding-agent 0.1a0, an experimental coding agent built on his LLM library. The agent provides tools for file editing, command execution, and code search, and is available v…
On June 1, 2026, GitHub Copilot switched from flat-rate requests to per-token billing, with each model call priced individually for input, cached-input, and output tokens. The change shifts pricing po…
GitHub Enterprise Cloud customers with enterprise managed users can now access Copilot agent session data in public preview, including prompts, responses, and tool calls, via a streaming endpoint or R…
Coinbase CEO Brian Armstrong revealed the company operates 1,200 full-time AI agents, restructuring engineering pods to include two to four humans and ten agents. Code shipped per developer has double…
A developer created a middle manager agent for an autonomous software factory that orchestrates Devin coding sessions, manages an issue tracker, and communicates via Slack. The agent operates continuo…
GitHub now allows enterprises to cap AI credit usage per cost center via the REST API, preventing one group from using credits paid for by another. The feature automatically calculates pool limits bas…
New AI coding benchmarks, including Snorkel AI's Senior SWE-Bench and Scale AI's SWE-Bench Pro, are replacing older evaluations like HumanEval and original SWE-bench to test senior-level engineering s…
Developer Yuri Koval recommends managing Claude Code configuration with a dotfiles repository and GNU Stow, enabling version control rollbacks and easy machine migration. The approach tracks skills, h…
A developer built a minimal control plane in Python that maintains a desired number of Docker containers, demonstrating level-triggered reconciliation. The system automatically restarted a killed cont…
CodeRabbit, the most installed AI app on GitHub and GitLab, has launched a new feature that shows the context behind each code review comment, allowing developers to see the source of a comment and di…
Software engineers are moving beyond single-agent autonomy ladders to a two-axis framework of agency and orchestration, as multi-agent systems become mainstream. Products like Claude Code and Codex no…
Cursor published CursorBench 3.1, a benchmark for AI coding agents, revealing Anthropic's Fable 5 models as top performers and Cursor's Composer 2.5 achieving near-top performance at 3% of the cost. G…
Building an in-house analytics MCP server for customer AI agents may seem cheap initially but incurs hidden costs in multi-tenant isolation, identity propagation, governed metrics, observability, and …
GitHub announced that enterprise administrators can now set Copilot to auto model selection by default for new conversations via the managed-settings.json file. The setting applies to users licensed t…
GitHub Models will be fully retired on July 30, 2026, affecting all customers including those with active usage. The playground, model catalog, inference API, and BYOK endpoints will be shut down, wit…
The Godot Foundation updated its contribution policy to ban autonomous AI agent use and AI-generated code, citing a surge in low-quality AI contributions that overwhelm reviewers and demoralize the co…
Port's experiment found that routing AI agents through a structured context lake instead of direct-to-MCPs cut token costs by 58%, and adding a skill file brought savings to 80%. The context lake pre-…
A malicious npm package called codexui-android stole OpenAI Codex authentication tokens from developers in a supply chain attack. The package, which had over 29,000 weekly downloads and was also bundl…
Developer pdumicz released Secret Shuttle, a local bridge that lets AI coding agents use secrets like API keys without ever seeing the plaintext values. The tool stores secrets in a local daemon vault…
Google Health launched a command-line interface (CLI) that lets users access their health data from devices like Fitbit Air and Pixel Watch, and build custom tools such as dashboards and automated ale…