Testland
Browse all skills & agents

gitleaks-scanning

Configures and runs gitleaks - Go-based secret scanner with `gitleaks git` (scan local git via `git log -p`), `gitleaks dir` (filesystem), `gitleaks stdin` (pipe); 100+ built-in rules + custom rules in `.gitleaks.toml` ([[rules]] with regex / entropy / keywords / tags); allowlist via [[rules.allowlists]] (commits / paths / stopwords); pre-commit hook + GitHub Action integration; baseline file for legacy debt. Use when the team needs OSS secret scanning at commit time + CI gate.

Install with skills.sh (any agent)

npx skills add testland/qa --skill gitleaks-scanning
View source

gitleaks-scanning

Command-API note (v8.19.0+): Per gl-gh (opens in new window), "v8.19.0 deprecated detect and protect commands, though they remain available." Current scanning modes:

CommandUse
gitleaks git <repo>Scan local git repository using git log -p
gitleaks dir <path>Scan directories and files (no git history)
gitleaks stdinStream data via pipe

Use git for git-aware scans (fastest, history-rewinding); dir for non-git (e.g., extracted CI artifact); stdin for diff-piping.

When to use

  • The team needs pre-commit + CI secret scanning.
  • A migration from a private-repo-history audit (catch secrets ever committed; not just current state).
  • Custom rule library for org-internal secret formats (internal API key prefixes, etc.).
  • Layered with trufflehog-scanning for live-validation cross-check.

How to use

  1. Install gitleaks (Step 1); confirm with gitleaks version.
  2. Run a full-history baseline scan: gitleaks git (Step 2).
  3. Add .gitleaks.toml with [extend] useDefault = true plus any org-internal custom rules (Step 3 - 4, references/custom-rules-and-triage.md).
  4. Triage findings: rotate real leaks, allowlist genuine false-positives with a Re-review-date (Step 5, references file).
  5. Wire the pre-commit hook for fast local feedback (Step 6).
  6. Add the CI job with fetch-depth: 0 as the catch-net gate (Step 7).
  7. On any confirmed leak, rotate the credential first, then scrub (Step 8).

Step 1 - Install

Per gl-gh (opens in new window):

# Homebrew
brew install gitleaks

# Docker
docker pull zricethezav/gitleaks:latest
docker run -v ${PWD}:/path zricethezav/gitleaks:latest git /path

# From source
git clone https://github.com/gitleaks/gitleaks.git
cd gitleaks
make build

Step 2 - Basic scans

# Scan current git repo (full history)
gitleaks git

# Scan a specific dir (no git)
gitleaks dir ./services/

# Stream + scan (e.g., scan a PR diff)
git diff main..HEAD | gitleaks stdin

# Output formats
gitleaks git --report-format json --report-path leaks.json
gitleaks git --report-format sarif --report-path leaks.sarif
gitleaks git --report-format csv --report-path leaks.csv

For PR-time scanning (faster - only check what changed):

gitleaks git --log-opts="origin/main..HEAD"

Step 3 - .gitleaks.toml config

Per gl-gh (opens in new window) config structure:

[[rules]]
id = "rule-identifier"
description = "rule description"
regex = '''regex-pattern'''
secretGroup = 3
entropy = 3.5
keywords = ["auth", "password"]
tags = ["tag1", "tag2"]

[[rules.allowlists]]
description = "ignore specific matches"
commits = ["commit-hash"]
paths = ['''file-path-regex''']
stopwords = ['''false-positive-term''']

Built-in rules cover AWS, GCP, Azure, GitHub, GitLab, Stripe, Twilio, Slack, npm, PyPI, etc. Use gitleaks <command> --no-banner to discover the full default rule list.

Step 4 - Custom rule example

Org-internal secret formats (internal API-key prefixes, etc.) go in [[rules]] blocks. Set [extend] useDefault = true to keep the built-in rules; without it, custom rules replace the defaults entirely. Full example (custom rule + per-rule and top-level allowlists) in references/custom-rules-and-triage.md.

Step 5 - False-positive triage (MANDATORY)

Suppress genuine false-positives with [[rules.allowlists]] (paths / commits), the top-level [[allowlists]], --baseline-path for legacy debt, or an inline # gitleaks:allow comment. Every allowlist entry needs a Re-review-date; audit them quarterly. The suppression priority table, the baseline workflow, and the justification template are in references/custom-rules-and-triage.md.

Step 6 - Pre-commit hook integration

Per gl-gh (opens in new window):

# .pre-commit-config.yaml
repos:
  - repo: https://github.com/gitleaks/gitleaks
    rev: v8.24.2
    hooks:
      - id: gitleaks

Pre-commit prevents commits containing secrets from being created. Faster local feedback than CI-only scanning.

Step 7 - CI integration

Per gl-gh (opens in new window):

# .github/workflows/gitleaks.yml
name: gitleaks
on: [pull_request, push]
jobs:
  scan:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v5
        with: { fetch-depth: 0 }   # full history needed for git scan
      - uses: gitleaks/gitleaks-action@v2
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          GITLEAKS_LICENSE: ${{ secrets.GITLEAKS_LICENSE }}   # for organizations

fetch-depth: 0 is critical - without full git history, git-aware scan only sees the latest commit.

Step 8 - Rotation when a secret is found

Finding a leaked secret in git history is not enough - git history is permanent (mirrored in clones, GitHub forks, archives). The secret IS exposed. Workflow:

  1. Rotate immediately - invalidate the leaked credential at the provider (AWS IAM key rotation, GitHub PAT revoke, etc.)
  2. Audit usage - check provider audit logs for unauthorized use during the exposure window
  3. Document the incident - track in postmortem (cross-ref post-mortem-author in the qa-process plugin)
  4. Add the leaked-pattern to gitleaks so future similar leaks are detected
  5. Optional: rewrite git history (BFG Repo-Cleaner / git filter-repo) - but assume the secret IS exposed regardless

For automated rotation workflow, see secrets-rotation-runner.

Worked example

A team enables gitleaks on a 4-year-old private repo.

  1. First full-history scan writes 3 hits: gitleaks git --report-format json --report-path leaks.json.
  2. Two are real - an AWS access key in commit abc1234 and a Slack token in a since-deleted config. Both are rotated at the provider (Step 8), and provider audit logs are checked for misuse.
  3. One is a dummy AWS key under tests/fixtures/aws.json. It gets a [[rules.allowlists]] path entry with an Approved-by and a Re-review-date (references file).
  4. The rotated-but-still-in-history hits are captured as a baseline (gitleaks git --report-path gitleaks-baseline.json), and CI runs --baseline-path gitleaks-baseline.json so only NEW leaks fail.
  5. A pre-commit hook (Step 6) and a fetch-depth: 0 CI job (Step 7) are added. The next PR that pastes a live Stripe key is blocked at commit time and again in CI.

Anti-patterns

Anti-patternWhy it failsFix
fetch-depth: 1 in CIgit-aware scan misses history; only catches new leaksAlways fetch-depth: 0 (Step 7)
Allowlist without Re-review-datePermanent debtMandatory template (Step 5)
Rely only on pre-commit (no CI)Bypass --no-verify; CI is the catch-netBoth pre-commit AND CI (Steps 6 - 7)
Skip baseline; legacy findings block all PRsTeam disables gitleaks--baseline-path (Step 5)
Find leak; assume git-history scrub fixes itLeaked secret IS exposed; assume compromiseRotate immediately (Step 8)

Limitations

  • Regex + entropy can't catch all secret formats; pair with trufflehog-scanning (live-validation) for higher precision.
  • Rewriting git history is destructive + only effective if all copies are scrubbed (forks / clones / mirrors typically aren't).
  • Custom rules require regex skill; complex secret formats (multi-line, base64-encoded) need careful authoring.
  • gitleaks itself doesn't rotate secrets; that's the secrets-rotation-runner workflow.

References

  • gl-gh (opens in new window) - repository, install, commands, config
  • gitleaks.io - landing page
  • trufflehog-scanning, kingfisher-scanning - sister scanners
  • secrets-rotation-runner - build-an-X for rotation workflow after detection

gitleaks custom rules and false-positive triage

View source (opens in new window)

gitleaks custom rules and false-positive triage

Custom org-internal rules and the mandatory allowlist / baseline / justification workflow. Referenced from Step 4 and Step 5 of the gitleaks-scanning skill.

Custom rule example

# .gitleaks.toml
[extend]
useDefault = true

[[rules]]
id = "internal-api-key"
description = "Internal API key (acme- prefix)"
regex = '''(?i)acme[_-]?api[_-]?key[_-]?[a-zA-Z0-9]{32}'''
secretGroup = 1
keywords = ["acme"]
tags = ["acme-internal"]

[[rules.allowlists]]
description = "Test fixtures"
paths = ['''tests/fixtures/.*\.json$''']

[[allowlists]]
description = "Project-wide allowlist (legacy commits)"
commits = ["abc1234", "def5678"]
paths = ['''vendor/.*''', '''third_party/.*''']

The [extend] useDefault = true keeps built-in rules; without it, your custom rules replace the defaults entirely.

False-positive triage (MANDATORY)

Suppression mechanisms in priority order:

MechanismWhereWhen to use
[[rules.allowlists]] paths.gitleaks.tomlPer-rule path exclusion (test fixtures, vendor)
[[rules.allowlists]] commits.gitleaks.tomlPer-rule commit exclusion (historical false positive)
[[allowlists]] paths.gitleaks.toml (top-level)All-rule path exclusion
--baseline-pathCI flagLegacy debt: only fail on NEW findings vs baseline
Inline # gitleaks:allow commentCodeSingle-line suppression

Baseline workflow (per gl-gh (opens in new window)):

# Create baseline
gitleaks git --report-path gitleaks-baseline.json

# Apply baseline (only new findings fail)
gitleaks git --baseline-path gitleaks-baseline.json --report-path findings.json

Justification template (mandatory in .gitleaks.toml):

[[rules.allowlists]]
description = """
Reason: tests/fixtures/* contains intentional dummy AWS credentials
        for SDK initialization tests; never used against real AWS.
Approved-by: alice@example.com
Re-review-date: 2026-09-15 (re-evaluate when SDK supports mock-mode injection)
"""
paths = ['''tests/fixtures/.*\.json$''']

Cadence: every quarter, audit .gitleaks.toml allowlist entries; expired re-review-date entries removed.

Related skills

kingfisher-scanning

Configures and runs Kingfisher for secret scanning with access mapping: discovers leaked credentials AND maps them to the IAM identities and cloud resources they expose (S3 buckets, RDS instances, etc.); Intel Hyperscan regex engine makes it the fastest option for large monorepos; 950 built-in rules (largest of the OSS scanners); multi-target (local files / Git history / GitHub / GitLab / AWS S3 / Docker images); live API validation plus offline checksum verification; suppression via `--skip-regex` / `--skip-word` / `--baseline-file` / inline `kingfisher:ignore`. Use when cloud-blast-radius context matters or scan time on a large repo is blocking. Front-loads access-mapping and Hyperscan speed to differentiate from trufflehog-scanning, which also does multi-target scanning and live validation but offers no IAM access mapping.

secrets-baseline-manager

Builds and maintains a unified secrets baseline/allowlist across gitleaks (.gitleaksignore + --baseline-path), TruffleHog (--results=verified filter + trufflehog:ignore), and Kingfisher (--baseline-file + --exclude/--skip-* flags); adopts legacy findings without blocking PRs; enforces a waiver lifecycle (expires + approved_by + reason) stored in .secrets-waivers.yaml; prevents baseline rot via quarterly audit + expiry enforcement. Use when onboarding secrets scanning onto a repo that already has historical findings, or when per-scanner ignore configs have drifted out of sync and need consolidating into one governed allowlist.

secrets-rotation-runner

Build-an-X for the secret-rotation workflow after detection - detect via gitleaks/trufflehog/kingfisher → identify provider via verifier → rotate via provider API (AWS IAM / GitHub PAT / Stripe / GCP / Azure / Twilio / Slack / etc.) → invalidate old secret → audit log via observability stack → post-mortem cross-ref. Use when a secret is detected in code (or proactively for periodic rotation) - assume git-history scrub does NOT prevent compromise.

trufflehog-scanning

Configures and runs TruffleHog v3 - secret scanner with **live verification** (validates discovered secrets against provider APIs to confirm actual exposure vs entropy false positive); supports per-source subcommands (`git`, `github`, `gitlab`, `filesystem`, `s3`, `docker`, `gcs`, `postman`); `--results=verified` filter for high-precision output; `--exclude-detectors=TYPE` for noise reduction; exits 183 on findings via `--fail`. Use when the team needs verified secret findings (low false-positive rate) or scans across cloud + repo + container surfaces.