This skill covers integrating Static Application Security Testing (SAST) tools—CodeQL and Semgrep—into GitHub Actions CI/CD pipelines. It addresses configuring automated code scanning on pull requests and pushes, tuning rules to reduce false positives, uploading SARIF results to GitHub Advanced Security, and establishing quality gates that block merges when high-severity vulnerabilities are detected.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
This skill covers integrating Static Application Security Testing (SAST) tools—CodeQL and Semgrep—into GitHub Actions CI/CD pipelines. It addresses configuring automated code scanning on pull requests and pushes, tuning rules to reduce false positives, uploading SARIF results to GitHub Advanced Security, and establishing quality gates that block merges when high-severity vulnerabilities are detected.
When development teams need automated code-level vulnerability detection on every pull request
When security teams require consistent SAST enforcement across all repositories in an organization
When migrating from manual or periodic security reviews to continuous security testing
When compliance frameworks (SOC 2, PCI DSS, NIST SSDF) require evidence of automated code analysis
When multiple languages coexist in a monorepo and need unified scanning under one workflow
Do not use for runtime vulnerability detection (use DAST instead), for scanning third-party dependencies (use SCA tools like Snyk), or for infrastructure-as-code scanning (use Checkov or tfsec).
Common Misconfigurations & Verification
SARIF upload mistaken for a gate:upload-sarif only populates the Security tab — it never fails the job. Make the gate real with semgrep ci --error (or --severity ERROR) and branch protection required_status_checks listing each scan job by name.
CodeQL autobuild skipping code: for compiled languages github/codeql-action/autobuild can build a subset and silently under-scan. Confirm the analyzed line count, or supply manual build steps; watch for "No source code was seen."
permissions: security-events: write missing: without it the SARIF upload fails and the PR still merges green. Set explicit job-level permissions and verify the upload step succeeded.
Over-suppression blind spots: broad paths-ignore globs or blanket nosemgrep lines can hide real sinks. Keep paths-ignore to tests/vendor only and audit suppressions against OWASP Top 10 / CWE Top 25.
PR-only scanning: running solely on pull_request never re-scans main against newly published rules. Add a schedule: cron full scan.
Verify by introducing a finding: open a draft PR adding a known sink (e.g. pickle.loads(user_input) or SQL string-format), confirm both CodeQL/Semgrep flag it at high severity and the required check goes red and blocks merge.
Prerequisites
GitHub repository with GitHub Actions enabled
GitHub Advanced Security license (required for CodeQL on private repos; free for public repos)
Semgrep account for managed rules and Semgrep App dashboard (free tier available)
Repository code in a supported language: Python, JavaScript/TypeScript, Java, C/C++, C#, Go, Ruby, Swift, Kotlin
Workflow
Step 1: Configure CodeQL Analysis Workflow
Create a CodeQL workflow that runs on pull requests and on a weekly schedule to catch vulnerabilities in existing code.
Semgrep complements CodeQL with faster scans and support for custom pattern-based rules. Configure it to upload SARIF results to the same GitHub Security tab.
# Example: Suppressing a known false positive in Semgrepimport subprocess
defrun_safe_command(cmd_list):
# nosemgrep: python.lang.security.audit.dangerous-subprocess-use
result = subprocess.run(cmd_list, capture_output=True, text=True, shell=False)
return result.stdout
Step 6: Aggregate and Report Findings
Use the GitHub Security Overview dashboard and configure notifications for security alerts across repositories.
# Query SARIF results via GitHub API for reporting
gh api repos/{owner}/{repo}/code-scanning/alerts \
--jq '.[] | select(.state=="open") | {rule: .rule.id, severity: .rule.security_severity_level, file: .most_recent_instance.location.path, line: .most_recent_instance.location.start_line}'# Count open alerts by severity
gh api repos/{owner}/{repo}/code-scanning/alerts \
--jq '[.[] | select(.state=="open")] | group_by(.rule.security_severity_level) | map({severity: .[0].rule.security_severity_level, count: length})'
Key Concepts
Term
Definition
SAST
Static Application Security Testing — analyzes source code without executing it to find security vulnerabilities
SARIF
Static Analysis Results Interchange Format — standardized JSON format for expressing results from static analysis tools
CodeQL
GitHub's semantic code analysis engine that treats code as data and queries it for vulnerability patterns
Semgrep
Lightweight static analysis tool using pattern matching to find bugs and security issues across many languages
Security Extended
CodeQL query suite that includes additional security queries beyond the default set for deeper analysis
Quality Gate
Automated checkpoint that blocks code from progressing through the pipeline unless security criteria are met
False Positive
A scan finding that incorrectly identifies secure code as vulnerable, requiring suppression or tuning
Tools & Systems
CodeQL: GitHub's semantic code analysis engine with deep dataflow and taint tracking analysis
Semgrep: Fast, lightweight pattern-matching SAST tool with 3000+ community rules and custom rule support
GitHub Advanced Security: Platform providing code scanning, secret scanning, and dependency review in GitHub
SARIF Viewer: VS Code extension for reviewing SARIF results locally during development
GitHub Security Overview: Organization-level dashboard aggregating security alerts across all repositories
Common Scenarios
Scenario: Monorepo with Multiple Languages Needs Unified SAST
Context: A platform team manages a monorepo containing Python microservices, TypeScript frontends, and Go infrastructure tools. Security reviews happen manually every quarter, missing vulnerabilities between reviews.
Approach:
Configure CodeQL with a matrix strategy covering Python, JavaScript, and Go languages
Add Semgrep with --config auto to detect language automatically and apply relevant rulesets
Create path-based triggers so only changed language directories trigger their respective scans
Upload all SARIF results to GitHub Security tab with unique categories per tool and language
Set branch protection requiring all SAST jobs to pass before merge
Schedule weekly full-repository scans to catch issues in unchanged code from newly published CVE patterns
Pitfalls: Setting CodeQL to analyze all languages on every PR increases CI time significantly. Use path filters to trigger only relevant language scans. Semgrep's --config auto may enable rules that conflict with CodeQL findings, creating duplicate alerts.
Scenario: Reducing Alert Fatigue from High False Positive Rate
Context: After enabling SAST, developers ignore findings because 40% are false positives, undermining the security program.
Approach:
Export all current alerts and categorize them as true positive, false positive, or informational
Create a custom CodeQL config excluding noisy query IDs that produce the most false positives
Write .semgrepignore patterns for test files, generated code, and vendored dependencies
Establish a weekly triage meeting where security and development leads review new rule additions
Track false positive rate as a metric and target below 15% for developer trust
Pitfalls: Over-suppressing rules to reduce noise can create blind spots. Always validate suppressions against the OWASP Top 10 and CWE Top 25 to ensure critical vulnerability classes remain covered.
Output Format
SAST Pipeline Scan Report
==========================
Repository: org/web-application
Branch: feature/user-auth-refactor
Scan Date: 2026-02-23
Commit: a1b2c3d4
CodeQL Results:
Language Queries Run Findings Critical High Medium
javascript 312 4 1 2 1
python 287 2 0 1 1
Semgrep Results:
Ruleset Rules Matched Findings Errors Warnings
auto 1,847 3 1 2
owasp-top-ten 186 2 1 1
custom-rules 12 1 0 1
QUALITY GATE: FAILED
Blocking findings: 2 Critical/High severity issues
- [CRITICAL] CWE-89: SQL Injection in src/api/users.py:47
- [HIGH] CWE-79: Cross-site Scripting in src/components/Search.tsx:123
Action Required: Fix blocking findings before merge is permitted.
