ln-761-secret-scanner（ln 761 秘密扫描仪）

Paths: File paths (shared/, references/, ../ln-*) are relative to skills repo root. If not found at CWD, locate this SKILL.md directory and go up one level for repo root. If shared/ is missing, fetch files via WebFetch from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}.

Secret Scanner

Type: L3 Worker Category: 7XX Bootstrap

Scans codebase for hardcoded secrets and credentials, returning structured findings for remediation.

Purpose & Scope

• Detect hardcoded secrets using available tools (gitleaks, trufflehog) or manual patterns
• Classify findings by severity (Critical/High/Medium/Low)
• Filter false positives via baseline and allowlists
• Provide remediation guidance per finding type
• Return normalized report to parent orchestrator (ln-760)

When to Use

• During project bootstrap (via ln-760-security-setup)
• Pre-commit hook validation
• CI/CD security pipeline
• Manual security audit

Workflow

Phase 1: Tool Detection

Step 1: Check Available Scanners

• Check if gitleaks is installed (preferred)
• Check if trufflehog is installed (alternative)
• If neither available: use manual pattern matching as fallback

Step 2: Load Configuration

• Load project .gitleaks.toml if exists (custom rules)
• Load .gitleaksbaseline if exists (known false positives)
• If no config: use default patterns from references/detection_patterns.md

Phase 2: Scan Execution

Step 1: Run Available Scanner

• Execute scanner against project root
• Capture output in structured format (JSON/SARIF preferred)
• If tool unavailable: run manual grep patterns for high-confidence secrets

Step 2: Parse Results

• Normalize output to common format: file, line, pattern, raw_match
• Preserve original severity if provided by tool

Phase 3: Report Generation

Step 1: Severity Classification

• Map findings to severity levels per references/detection_patterns.md
• Critical: AWS keys, private keys, JWT secrets
• High: Generic passwords, connection strings
• Medium: API keys (may be test data)
• Low: Potential secrets requiring manual review

Step 2: False Positive Filtering

• Apply baseline exclusions
• Apply allowlist patterns (placeholders, test data, docs)
• Mark filtered items as "excluded" with reason

Step 3: Build Report

• Group findings by severity
• Include file path, line number, pattern matched
• Do NOT include actual secret values in report

Phase 4: Remediation Guidance

Step 1: Attach Remediation Actions

• For each finding, attach remediation steps from references/remediation_guide.md
• For Critical findings: emphasize immediate rotation requirement

Step 2: Return Results

• Return structured findings list to orchestrator
• Include summary: total scanned, total found, by severity

Critical Rules

1. Never log actual secret values - redact in all outputs
2. Treat any found secret as compromised - rotation required for Critical
3. Preserve baseline - do not remove existing baseline entries
4. Pre-commit priority - recommend pre-commit hook if not configured
5. Git history awareness - warn if secret may exist in history (requires git-filter-repo)

Definition of Done

• Scan completed using available tool or manual patterns
• Findings classified by severity
• False positives filtered via baseline/allowlist
• Remediation guidance attached to each finding
• Report returned in normalized format (no raw secret values)
• Critical findings flagged with rotation requirement

Reference Files

Version: 2.0.0 Last Updated: 2026-01-10