Spec-Driven Development

Practitioners reported coding agents following outdated agents.md files and generic project instructions. Other posts tied long-context drift and weak domain language to cleanup-heavy coding runs.

Matt Pocock argued teams should delete docs that restate code and keep ADRs, glossaries, and thin navigation pointers. Zeeg reported colocated repo specs drifted and did not steer agents enough.

Microsoft open-sourced SkillOpt, a system that treats agent skill documents as tunable artifacts and improves them against measured task batches. It matters because practitioners are already standardizing shared /research, QA, and packageable skills across harnesses, turning skill files into a new optimization surface alongside models.

Two days after Factory 2.0 launched, Factory added AutoWiki, which turns a repo into structured engineering docs that update on every push and can be generated from Droid with /wiki. The feature keeps architecture and request-path context aligned with the live codebase.

Practitioners published tests-first coding-agent workflows built around red-green TDD, Hurl suites, GitHub label actions, and Codex-based execution checks. The pattern matters because verification remains the main bottleneck once generation is fast, especially in longer multi-file sessions.

A day after HTML artifacts surfaced as a Claude Code workflow, Anthropic promoted a `frontend-slides` plugin with direct install commands and artifact publishing. The rollout sharpened a real workflow split: teams are using HTML for human review and demos, while keeping markdown or MDX for token-efficient agent context.

CopilotKit shipped an interactive A2UI Composer for building widgets, inspecting AG-UI event streams, and generating reusable A2UI JSON. Teams can now prototype and copy agent-facing UI components without hand-authoring every widget schema.

Bram Cohen used the Claude Code leak to argue that prompt-only development produces bad software, while a separate 250-hour syntaqlite build said the durable version arrived only after a Python-to-Rust rewrite. Practitioners say specs, tests, linters, repo skills, and codebase context are the controls that keep coding agents maintainable.

Anthropic is testing a new /init flow that interviews users and configures Claude.md, hooks, and skills in new or existing repos. Try it in a sandbox repo, then watch for skills behavior differences between chat and web surfaces.

ACE open-sources a platform that turns AGENTS.md instructions into evolving playbooks backed by execution history, with hosted and self-hosted options. It is a notable response to prompt drift and prompt extraction, because procedures become revisable operating docs instead of static prompts.

Google rolled out a redesigned Stitch workspace that accepts text, code, PRDs, and images on a spatial canvas, then generates prototypes and portable DESIGN.md files. Teams testing AI-native UI workflows can use it to try a tighter design-to-code loop in the live product.

Geoffrey Huntley published a four-loop Ralph workflow for porting codebases by turning tests and source into cited specs before implementation. Try it when you need AI help translating a mature codebase across languages without losing behavioral coverage.

OpenAI detailed how repo-local skills, AGENTS.md, and GitHub Actions now drive repeatable verification, release, and pull request workflows across its Agents SDK repositories. Maintainers can copy the pattern to reduce prompt sprawl and keep agent behavior closer to the codebase.