Urgent.News

What's breaking now, across thousands of outlets.

Sunday, 6 September 2026 Reading Room

AI

15 AI Coding Prompts for Codex and Claude Code

Code audits, performance work, feature planning, and product research, with an installable workflow skill. Theo shared six ways he uses Astra : code cleanup, performance work, agent setup, PR and issue triage, merging, and taking over stalled work. I wrote a prompt for each and added nine for other development tasks, including feature planning, competitor research, and SEO/GEO audits. You can…

dev.to ·

The report provides a set of 15 AI coding prompts for use with Codex and Claude Code, which can be found on GitHub or installed as an Astra Workflows skill. These prompts cover various development tasks such as code cleanup, performance work, feature planning, and SEO/GEO audits. To use these prompts, one should follow the shared instructions, fill in the bracketed fields, and review the permissions for each prompt.

It is recommended to start a new session for each objective and keep related tasks on separate branches or worktrees with clear ownership. The prompts should be adapted to the specific project's AGENTS.md or CLAUDE.md files, and users should run one task at a time within a session, connecting the selected model and its tools in the agent setup.

Brief written by urgent.news from Dev.to's own syndicated text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Ten fine-tuning mistakes I see students make (and made myself)

I built FineTune Studio after making most of these mistakes on my own runs. They are not exotic. They are the difference between "I fine-tuned a model" on a resume and "I fine-tuned a model and here…

  • Always start with a baseline model to compare fine-tuning results.
  • Use validated data for training, not unvalidated data, to avoid errors.
  • Define evaluation metrics and held-out sets before training to guide experimentation.

·dev.to

Running ML models in the browser with ONNX Runtime Web: a practical guide

FaceVision does face detection, recognition and liveness checks without sending a single frame to a server, because every model runs in the browser through ONNX Runtime Web.

  • FaceVision runs ML models in the browser without sending data to a server
  • ONNX Runtime Web facilitates model execution in the browser
  • Privacy-focused design stores only vector embeddings in FastAPI and PostgreSQL

·dev.to

More from Sunday 6 September →