Repository navigation
Quick Start
Copy-paste workflows for the most common tasks. Each assumes the binary is installed and the libraries are set up (prowl rules update / prowl verifiers update; see Installation).
Walk the current directory and report findings. Binaries, large files, and noise dirs (.git, node_modules, vendor, ...) are skipped automatically.
prowl scan .Scan exactly what is staged for commit — fast, and the natural pre-commit hook. Exits non-zero only if a finding meets the --fail-on threshold.
prowl scan --staged --fail-on highScan files changed since a revision (any git ref: a branch, tag, or SHA). Ideal for CI on pull requests.
prowl scan --since origin/main --fail-on mediumDrop everything except secrets the provider confirms are still active — the strongest false-positive filter.
prowl scan . --verified-onlyEmit a SARIF report to a file for GitHub code scanning or any SARIF-aware tool.
prowl scan . --format sarif -o prowl.sarifPrint the report and exit 1 if any finding is at least high, failing the pipeline. Combine with -s to suppress the report and use only the exit code.
prowl scan . --fail-on high # report + non-zero exit on high+
prowl scan . --fail-on high -s # exit code only, no outputSnapshot today's findings as accepted, then on later runs suppress everything in the baseline so only new secrets surface.
prowl scan . --write-baseline .prowl-baseline.json # record current state, exit 0
prowl scan . --baseline .prowl-baseline.json # report only new findingsA .prowl-baseline.json in the working directory is picked up automatically, so once it exists prowl scan . already suppresses known findings.
Stream every blob ever committed to catch secrets that were removed from HEAD but still live in history.
prowl scan --history --format json -o history.jsonGetting started
Scanning
- Scanning Files
- Repository Scanning
- Org Scanning
- Container Scanning
- Bucket Scanning
- Domain Scanning
- Jira & Confluence Scanning
Detection & verification
Integration
Reference