Most people only use Claude Code interactively. But the same CLI has a headless mode: claude -p "<prompt>" runs one prompt, prints the answer, and exits. That makes it a normal script step you can drop into CI or a git hook.

The one line 

git diff origin/main...HEAD | claude -p "Review this diff. List real bugs only. If there are none, print exactly: LGTM" --output-format text

Piped stdin becomes context, -p runs non-interactively, and the output is plain text you can grep.

Make it fail the build 

#!/usr/bin/env bash set -euo pipefail out=$(git diff origin/main...HEAD | claude -p \ "Review this diff for bugs, missing error handling and leaked secrets. If nothing is wrong, reply with exactly LGTM." \ --allowedTools "Read,Grep" --max-turns 3) echo "$out" [[ "$out" == "LGTM" ]] || exit 1
  • --allowedTools "Read,Grep" keeps the CI run read-only: it can look at files but not edit or run shell commands.
  • --max-turns 3 caps how long it can wander.
  • Need structured results? Use --output-format json and parse .result with jq.

GitHub Actions step 

- name: Claude review env: ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }} run: | npm i -g @anthropic-ai/claude-code ./scripts/claude-review.sh

Keep the key in repository secrets, never in the script.

Tips

  • Keep the prompt strict ("bugs only", "exactly LGTM") so the output is easy to check.
  • Your project CLAUDE.md and .claude/settings.json still load in headless mode, so your house rules apply in CI too.
  • Start as a non-blocking job (continue-on-error: true) for a week before you let it fail builds.

Want the rest of the setup?

I keep copy-ready Claude Code configs (CLAUDE.md templates, slash commands, hooks, permission rules) in Vildanden Agent Config. There's a free sample: https://justly-mia-429.rehost.page/