Tag
AI coding agents
AI coding agents are agentic tools that work in terminals, editors, or CI to write code, edit files, run tests, and open PRs. They matter because they can speed up delivery, but also raise questions about permissions, token cost, regressions, and real-world reliability.
6 articles
Why AI-agent CLIs are the new supply-chain attack surface
AI-agent CLIs create a new supply-chain backdoor class that scanners do not catch.
Why AI coding agents need an architecture compiler
Atomadic Forge proves AI code needs structural enforcement, not just tests and linting.
From Prompting to Harness Engineering
OpenAI says one team shipped a 1M-line product with 3 engineers and Codex, merging about 1,500 PRs in 5 months.
I Tested Devin on 10 Tasks. It Finished 3.
Devin scored 13.86% on SWE-bench and finished 3 of 10 real tasks in one test, showing where AI coding agents still fall short.
Claude Code Setup Guide for Research Workflows
A practical setup guide for Claude Code in research workflows, with terminal tips, context-window advice, and pricing details.
RTK cuts Claude Code token spend fast
RTK claims it can cut Claude Code token use by up to 80% by routing work through local shell commands and agents.