Use --output-format for Structured JSON Responses
Need Claude to return data in a format you can pipe into another tool? The --output-format flag forces structured JSON output.
claude -p "list all TODO comments in this project" --output-format json
Instead of a conversational response, you get clean, parseable JSON that you can feed directly into jq, a script, or another command.
This is essential for building automation on top of Claude Code. A few examples:
# Extract function names and pipe to another tool
claude -p "list all exported functions in src/utils/" \
--output-format json | jq '.result'
# Generate a structured changelog
claude -p "summarise changes since last tag as JSON with
fields: type, scope, description" --output-format json
# Chain Claude into a shell pipeline
claude -p "analyse this error log" --output-format json \
< /var/log/app.log | jq '.suggestions[]'
Combine this with the -p (print/pipe) flag for non-interactive usage. Together they turn Claude Code from a conversational tool into a programmable one.
# Non-interactive + structured output = scriptable Claude
claude -p "what dependencies are outdated?" \
--output-format json > outdated.json
When you need data, not conversation — --output-format json makes Claude Code scriptable.
Log in to leave a comment.
Set up Claude Code as an automated reviewer in your CI pipeline — on every pull request, it reads the diff, checks for bugs, security issues, missing tests, and convention violations, then posts its findings as a PR comment. Your human reviewers get a head start because the obvious issues are already flagged before they look.
Before deploying, tell Claude to read your project — migrations, environment variables, queue workers, scheduled tasks, caching, third-party integrations — and generate a deployment checklist that's specific to your app. Not a generic "did you run migrations?" list, but one that knows YOUR infrastructure and catches the things YOUR deploy can break.
Instead of writing a README from memory or copying a template, tell Claude to read your project and generate one that's actually accurate — real setup instructions from your config, real architecture from your directory structure, real API examples from your routes, and real prerequisites from your dependency files.