alperduzgun/Warden Security Scan

AI-native security and quality gate for LLM-generated code

Trust Signals

Scorecard Score: not yet scored
Maintenance Recency: Activelast commit Apr 8, 2026
License: Apache 2.0

Pinned Snippet

workflow.ymlSHA-pinned

uses: alperduzgun/warden-core@556d574f16a9fe9fe356334c8742a1425225afde # v3.0.0

tags can be moved; commit SHAs can't. why a SHA?

Inputs

name	description	required	default
scan-path	Path to scan (default: current directory)	no	.
level	Analysis level: basic, standard, deep	no	standard
format	Output format: sarif, json, text, markdown	no	sarif
output-file	Output file path	no	warden.sarif
quick-start	Run deterministic-only scan (no LLM required)	no	false
python-version	Python version to use	no	3.11
diff-mode	Scan only changed files in this PR (faster, recommended for PR gates)	no	false
extra-args	Additional arguments passed to warden scan	no	""
upload-sarif	Upload SARIF to GitHub Security tab	no	true
post-comment	Post scan results as a PR comment (idempotent, updates existing)	no	true
fail-on-severity	Fail workflow if findings at this severity or above: critical, high, medium, low, none	no	critical
github-token	GitHub token for PR comments and SARIF upload	no	${{ github.token }}

Outputs

name	description
sarif-file	Path to generated SARIF report
exit-code	Scan exit code (0=pass, 1=infra error, 2=policy violation)
findings-count	Total number of findings
quality-score	Quality score (0.0-10.0)