Spec-first is an open-source, MIT-licensed plugin and workflow for developing software with AI agents. Its central argument is that agents should agree on a checked specification before writing code, replacing loosely directed “vibe coding” with a repeatable path from an idea to a verified pull request. The user describes a feature; the agent first reads project files and stored preferences, then creates a YAML spec with the goal, evidence from specific files and lines, implementation steps, and completion criteria. A separate reviewer, without the original conversation, checks the spec against the code and may cut unsupported work or add missing requirements. Only then does an agent implement the plan and its tests.
Each spec gets its own branch and Git worktree, allowing independent tasks to proceed in parallel. The workflow records non-obvious design choices, tracks resumable state, and runs declared build and test checks before allowing a commit; failed checks block delivery. A successful run produces a commit and pull request containing the spec. Larger efforts can be divided into dependent or independent specs, with compatible tasks built concurrently and the merged result verified. The plugin supports Claude Code directly, while other agents can use the same repository files and a Git hook. Its creators say the method was used to build Agent Smith and report more than 1,300 completed specs, 1,190 decision records, and 650 specs that passed the gate.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.
Amazon updated its software engineer guidelines to emphasize results over specific tools, including AI. It says engineers must combine AI with human judgment and expertise while adapting to rapid changes in software development. (aboutamazon.com)
Trump said he would ban the term “artificial intelligence” in official documents and other materials. He wants it replaced with “super intelligence,” which he calls a more accurate name. (reuters.com)
Researchers found that pushing individual neurons in Qwen3-8B led it to identify as 34,220 different personas, rather than as an AI. The results come from one prompt, and the persona labels were assigned by an AI judge without human review. (qwen-neurons-site.vercel.app)
New records show Medicare’s WISeR AI prior-authorization program has caused care delays, denials, operational problems, and reported patient harm. The EFF says the findings highlight the need for stronger legal limits on government use of automated decisions. (eff.org)
An Anthropic hiring manager has dropped coding interviews because AI can outperform candidates on those tests. Junior engineers can still excel by using AI, questioning assumptions, and bringing fresh perspectives; hunger can outweigh tenure. (felixrieseberg.com)
Anthropic expanded its GSA agreement, offering government agencies a 40% discount on Claude products and implementation support, with no minimum spending requirement. The deal extends to state, local, tribal, and territorial governments and is estimated to save the government $1 billion. (anthropic.com)
Today's best Hacker News stories, summarized and screenshotted, one email a day.