S
Sprint Log Parser
Docs / Intrusion Detection / Threat Vector Parsing

Threat Vector Parsing

Understand the signature scanning rules and internal matching logic utilized by Sprint Log Parser to identify web intrusion patterns.

1. Signature Rules

The ingestion parser uses an advanced regex-based heuristics engine to check every single incoming line (URI path, query string, request body, and user agent). Each incoming line is parsed against predefined expressions corresponding to known web attack patterns.

Signatures are optimized for speed, performing multi-stage string matches in microseconds before evaluating heavy regular expressions.


2. Common Attack Vectors

Sprint Log Parser classifies attacks into specific hazard categories:

Threat Category Detection Rule Pattern Severity
SQL Injection (SQLi) SELECT.*FROM|UNION.*SELECT|OR.*1=1 Critical
Cross-Site Scripting (XSS) <script>|javascript:|onerror= High
Path Traversal (LFI) \.\./\.\./|/etc/passwd High
Remote Code Execution (RCE) shell_exec|curl.*\|.*bash|eval\( Critical

3. Real-time Threat Alerts

When a log entry triggers one of the vector rules, the system immediately registers a `ThreatAlert` record associated with the project.

Alert dispatches are queued instantly using background jobs, ensuring that configured webhook channels (like Slack Block Kit notifications) and recipient email lists receive details about the attack (attacker IP, matched rule, payload snippet, and geographical origin) within seconds.