json-validation

JSON validation defends against injection and parsing vulnerabilities by enforcing strict schema compliance. This Skill centralizes validation logic to catch malicious or malformed input before it reaches agent decisioning logic, ensuring safe, consistent parsing across all agents.

Yes. This Skill uses jq to perform defensive JSON parsing and strict schema enforcement on AGENT_SUCCESS_CRITERIA. It exposes utilities like validate_success_criteria and validate_criteria_structure to verify structure, extract test suite data, and handle missing fields gracefully.

Use the exported helper functions: get_test_suite retrieves suite metadata, get_test_command extracts per-suite commands, and get_pass_threshold retrieves numeric thresholds. These utilities parse JSON safely while preventing injection attacks.

The Skill implements graceful handling of missing fields and provides secure error messaging. Invalid structure is caught during validation; validate_criteria_structure confirms compliance before use, preventing downstream failures in test orchestration.

Yes. Validation is applied at startup and during test orchestration across all 21 CFN Loop agents. The centralized approach ensures consistent, secure parsing of AGENT_SUCCESS_CRITERIA regardless of which agent executes the criteria.

Centralized validation eliminates parsing inconsistencies and reduces injection vulnerability surface area. A single source of truth for schema enforcement, error handling, and secure messaging scales safely across multiple agents and prevents duplicate vulnerability logic.