sarif-parsing

This skill acts as a specialized assistant for the SARIF (Static Analysis Results Interchange Format) 2.1.0 standard, providing the capability to interpret, filter, and transform security scan data generated by industry-standard tools such as CodeQL, Semgrep, and various vulnerability scanners. It is designed for security engineers, DevOps professionals, and developers who need to manage large-scale security findings programmatically. Instead of manually inspecting output, users can utilize this skill to automate the ingestion and analysis of static analysis artifacts, facilitating a cleaner and more actionable security feedback loop within CI/CD pipelines.

• Aggregates findings from multiple fragmented SARIF files into a unified dataset for comprehensive vulnerability tracking.

• Performs intelligent deduplication of alerts to eliminate noise and reduce false positives across multiple scan runs.

• Filters security results based on specific severity levels, rule IDs, or file paths using powerful CLI-based query tools like jq or advanced Python libraries.

• Converts SARIF outputs into human-readable report formats such as CSV, HTML, or custom JSON structures for dashboard integration.

• Provides programmatic access to SARIF data models using Python libraries like pysarif and sarif-tools, enabling custom analysis workflows.

• Calculates baseline comparisons to detect regression and verify whether a finding is new or has been previously suppressed.

• Use this skill to parse and interpret existing SARIF output files; it does NOT trigger new security scans.

• To run new scans, utilize the respective CodeQL or Semgrep skills in the marketplace.

• Typical inputs include .sarif or .json files containing scan results; typical outputs include filtered reports, summary statistics, or modified result files.

• The skill provides deep support for the OASIS SARIF 2.1.0 standard, ensuring full compatibility with tool-agnostic data processing.

• Ideal for use cases involving multi-tool security pipelines where normalization of disparate scan formats is required to maintain a single source of truth.