Run agents through quality-gated build, test and deploy tools
The author has agents use managed tools that enforce coverage, dependency and lint metrics, build, test and deploy in about 10s, and pick up daily quality-check findings as work. ·
Done with a custom-built or unnamed agent
- What they did
- The author gives agents managed tools that enforce quality metrics (test coverage, allowed dependencies, linters), then build, test and deploy roughly in parallel and prepare the work for git. The tools run on the deployment machine for speed and submit attestations of everything they do. A daily process runs further checks, such as outdated dependencies and slow tests, and turns findings into work for agents.
- How it went
- The full cycle, including docker build and deploy to a test environment, takes 10 seconds or less. The author's own review is mostly checking the app does what it should, and passing metrics matter more than reading the code.
- Worth knowing
- The author built their own system to track the attestations, and finds a 10s budget achievable on every project they've worked on.
Try it yourself with Claude Code
In my repo at [path or repo name], set up quality gates for [coverage threshold, lint rules, dependency checks] and a single command that builds, tests and deploys to [environment]. Then run the daily quality checks, turn each finding into a task list, and fix the ones that are low risk. Stop when the gates pass and the build and test command runs cleanly, and pause to show me the plan before any real deploy or deletion.
Want help getting this working? Join the Vibe Coding Academy ↗ Run by Alex Finn, who also runs this site.
Source: HN · Nov 13, 2026
Related use cases
Build a C compiler with a team of parallel agents
Anthropic's engineering write-up on tasking Opus 4.6 agent teams with building a C compiler.
Build an iCloud skill with a Muse agent
A screen-reader user asked their Muse agent to build an iCloud skill and it produced something, but accessibility problems with the text field and conversation navigation got in the way.
Resolve a Slack-reported issue with a coding agent
A Codex agent resolved an issue passed to it in about 40 minutes, and an ops agent verified the fix back in Slack.
Run terminal commands with Copilot agent mode
Users report the agent sometimes fails to read terminal output and may undo correct code; workarounds include switching shells.
Migrate a large web application with Copilot Agents
A developer found Copilot Chat struggled with a large web app migration, so switched to Copilot Agents, which handled the codebase better; the migration was still early.
Catch bugs by exploring web and mobile apps with QA agents
Momentic's agents autonomously plan, explore and bug-bash user flows in web, iOS and Android apps for engineering teams.
Custom agent
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.