Install any skill in seconds. Free to start, no credit card required.
Get Started Free →30+ service-specific secret detection regex patterns, entropy-based detection, PEM/JWT/Base64 identification, and false positive filtering.
.claude/skills/secret-patterns/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 100% | 11 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✗→✓ | ▲ Improved | — | — |
| case-17 | ✗→✓ | ▲ Improved | — | — |
| case-01 | ✗→✓ | ▲ Improved | — | — |
| case-03 | ✗→✓ | ▲ Improved | — | — |
| case-14 | ✗→✓ | ▲ Improved | — | — |
Patterns for finding leaked credentials in codebases, git history, and CI logs.
bash# Access Key ID: always starts with AKIA (long-term) or ASIA (session) AKIA[0-9A-Z]{16} ASIA[0-9A-Z]{16} # Secret Access Key: 40-char base64-ish string after aws_secret aws_secret_access_key\s*=\s*[A-Za-z0-9/+=]{40} # ripgrep one-liner rg --no-heading -n '(AKIA|ASIA)[0-9A-Z]{16}' .
bash# Personal access tokens (classic and fine-grained) ghp_[A-Za-z0-9]{36} github_pat_[A-Za-z0-9_]{82} # OAuth / app tokens gho_[A-Za-z0-9]{36} ghs_[A-Za-z0-9]{36} ghu_[A-Za-z0-9]{36} ghr_[A-Za-z0-9]{36} rg --no-heading -n 'gh[pousr]_[A-Za-z0-9]{36}' .
bash# Live secret (never commit) sk_live_[A-Za-z0-9]{24,} # Test secret (flag but lower severity) sk_test_[A-Za-z0-9]{24,} # Publishable keys (public, lower severity) pk_live_[A-Za-z0-9]{24,} pk_test_[A-Za-z0-9]{24,} rg --no-heading -n 'sk_(live|test)_[A-Za-z0-9]{24,}' .
bash# OpenAI sk-proj-[A-Za-z0-9\-_]{50,} sk-[A-Za-z0-9]{48} # Anthropic sk-ant-[A-Za-z0-9\-_]{90,} rg --no-heading -n '(sk-proj-|sk-ant-)' .
bash# Three base64url segments separated by dots eyJ[A-Za-z0-9\-_]+\.eyJ[A-Za-z0-9\-_]+\.[A-Za-z0-9\-_]+ # Decode header to verify (Python) import base64, json header = token.split('.')[0] + '==' print(json.loads(base64.urlsafe_b64decode(header)))
bash# RSA, EC, OpenSSH private keys -----BEGIN (RSA |EC |OPENSSH )?PRIVATE KEY----- rg --no-heading -n '\-\-\-\-\-BEGIN.+PRIVATE KEY' . # Also catch generic PEM blocks -----BEGIN CERTIFICATE----- -----BEGIN PGP PRIVATE KEY BLOCK-----
bashxoxb-[0-9]{11,}-[0-9]{11,}-[A-Za-z0-9]{24} # Bot token xoxp-[0-9]{11,}-[0-9]{11,}-[A-Za-z0-9]{32} # User token xoxe\.[A-Za-z0-9\-_]{80,} # Enterprise grid xoxs-[A-Za-z0-9\-]{80,} # SCIM token rg --no-heading -n 'xox[bpse][-.]' .
bash# PostgreSQL postgres(ql)?://[^:]+:[^@]+@[^/]+/\S+ postgresql\+asyncpg://[^:]+:[^@]+@ # MongoDB mongodb(\+srv)?://[^:]+:[^@]+@ # MySQL / MariaDB mysql://[^:]+:[^@]+@ rg --no-heading -n '(postgres|mongodb|mysql)://\S+:\S+@' .
bash# NPM publish token npm_[A-Za-z0-9]{36} # SendGrid SG\.[A-Za-z0-9\-_]{22}\.[A-Za-z0-9\-_]{43} # Twilio account SID + auth token AC[0-9a-f]{32} # Account SID SK[0-9a-f]{32} # API Key SID # Mailgun API key key-[A-Za-z0-9]{32} rg --no-heading -n '(npm_[A-Za-z0-9]{36}|SG\.[A-Za-z0-9_-]{22}\.)' .
bash-----BEGIN OPENSSH PRIVATE KEY----- -----BEGIN RSA PRIVATE KEY----- -----BEGIN EC PRIVATE KEY----- -----BEGIN DSA PRIVATE KEY-----
typescript// Detect the "type": "service_account" JSON pattern rg --no-heading -n '"type"\s*:\s*"service_account"' . // Or the private_key field rg --no-heading -n '"private_key"\s*:\s*"-----BEGIN' .
pythonimport math from collections import Counter def shannon_entropy(s: str) -> float: if not s: return 0.0 freq = Counter(s) length = len(s) return -sum((c / length) * math.log2(c / length) for c in freq.values()) def is_high_entropy_secret(value: str, threshold: float = 4.5) -> bool: # Ignore short strings and common words if len(value) < 20: return False return shannon_entropy(value) > threshold # Example usage candidates = [ "AKIAIOSFODNN7EXAMPLE", # entropy ~4.1 (low, example key) "wJalrXUtnFEMI/K7MDENG/bPxRfi", # entropy ~4.7 (real secret) "my-password-123", # entropy ~3.8 (low) ] for c in candidates: print(f"{c[:20]}... entropy={shannon_entropy(c):.2f} secret={is_high_entropy_secret(c)}")
pythonFALSE_POSITIVE_PATHS = [ r'test[s]?/', r'__tests__/', r'spec/', r'fixtures/', r'examples/', r'docs/', r'\.md$', r'CHANGELOG', r'README', ] FALSE_POSITIVE_VALUES = [ 'example', 'placeholder', 'your_key_here', 'INSERT_KEY', 'REPLACE_ME', 'xxxx', '1234', 'test', 'dummy', 'fake', ] def is_false_positive(path: str, value: str) -> bool: import re for pattern in FALSE_POSITIVE_PATHS: if re.search(pattern, path, re.IGNORECASE): return True lower = value.lower() return any(fp in lower for fp in FALSE_POSITIVE_VALUES)
yaml# .pre-commit-config.yaml repos: - repo: https://github.com/gitleaks/gitleaks rev: v8.18.4 hooks: - id: gitleaks - repo: https://github.com/trufflesecurity/trufflehog rev: v3.63.5 hooks: - id: trufflehog entry: trufflehog git file://. --since-commit HEAD --only-verified --fail
bash# Manual scan of full git history gitleaks detect --source . --log-opts="--all" # TruffleHog targeted scan trufflehog git https://github.com/org/repo --only-verified # ripgrep composite scan (fast, no install) rg --no-heading -n \ '(AKIA|ASIA|ghp_|sk_live_|sk-proj-|sk-ant-|xoxb-|npm_|SG\.|-----BEGIN.+PRIVATE)' \ --glob '!{node_modules,dist,.git}' .
Key rule: Any match outside test/, docs/, or examples/ with entropy > 4.5 is a confirmed finding. Rotate immediately.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-13 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-10 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-20 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-04 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-06 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-18 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-15 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-11 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-22 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-17 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-01 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-16 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-19 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-08 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-09 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-05 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-07 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-12 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-03 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-02 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-14 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +23 percentage points is the difference between those two pass rates over the 21 comparable cases.
The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.
Other measured skills in the registry, with their headline benchmark lift.