A collection of 54 posts
Use Claude Code’s ! shell commands to verify reality without leaving the session: run the exact failing command, inspect logs, check git state, and feed the evidence back before the second pass.
Read more
Use @file to anchor Claude Code on the exact source of truth before it edits: read the file, state the hypothesis, patch narrowly, then verify with tests and git diff.
Read more
Use /memory to turn repeated Claude Code corrections into scoped project rules: save durable conventions with evidence, verification commands, and clear boundaries instead of vague preferences.
Read more
Use @folder to give Claude Code a bounded subsystem: map the files, identify the active path, edit narrowly, and verify with targeted tests and git diff.
Read more
A field guide to using Claude Code for frontend fixes: make it reproduce the bug, inspect the DOM, verify in the browser, and correct the patch before declaring done.
Read more
Use Claude Code to debug deployments from real logs: capture the first failing line, form a narrow theory, patch only the proven cause, redeploy, and verify the next run.
Read more
A field guide to making Claude Code name the exact failure signal before it edits: reproduce what is broken, define the proof, patch narrowly, and verify the same signal changed.
Read more
Use Claude Code more safely by keeping a visible acceptance checklist: define proof before edits, update it after each command, and refuse to close the session until every item has evidence.
Read more
Use Claude Code to turn a vague GitHub issue into a verified workplan: read the linked code, split the task into checks, expose missing assumptions, and create a first patch only after the plan survives inspection.
Read more
Use Claude Code for test-driven debugging: make it reproduce the failure first, inspect the exact failing assertion or log, apply a narrow fix, then run the same test again before broadening scope.
Read more