DEV Community

tdk-cli
tdk-cli

Posted on

TDK CLI Core for Teams: Drift Checks, Dry Runs, and Readiness

A local stack that works on one laptop is a start. A local stack that works on every laptop on the team is what saves time. The gap between them is usually drift: a file edited here, a port changed there, a service started by hand on one machine. TDK CLI Core addresses that gap with three features a team can rely on.

Drift checks

TDK records a snapshot of each generated file when it writes it. Before the stack starts, tdk up compares the current files with those snapshots. If someone edited a generated file by hand, the start stops and names the file. The fix is to change the service.json or the generator, and regenerate.

tdk config verify extends the same idea to the wider project. It checks the discovered service.json files and compares the generated service files with their snapshots.

Dry runs in review

tdk up main --dry-run shows the plan without starting anything: which services run, their ports and health URLs, and the Tilt command. A reviewer can run it on a branch and see the effect of a change to a service.json without starting containers.

Readiness, not guesses

tdk status --json reports readiness, with ready, pending, and failures, so a script or a teammate can wait for a clear answer. tdk up --json reports ok: false with errors when a resource fails or readiness times out.

A team routine

  1. Commit service.json files, and never commit generated output.
  2. Run the drift check in CI, so a hand-edited generated file fails the build.
  3. Use --dry-run in code review when a service's dependencies or ports change.
  4. Put the one-command startup in the project's README, and test it on a clean machine.
  5. Agree on shared port and naming conventions, so services find each other the same way everywhere.

Limits

These features make the local environment consistent. They do not make the application correct, and they do not replace production checks. TDK CLI Core is for development and testing. Production deployment stays with the team's existing tooling.

Summary

Drift checks catch hand edits, dry runs make changes reviewable, and readiness output replaces guesswork. Together they turn a local stack from a personal setup into one the whole team can start the same way.

Top comments (0)