shift-left-testing

Pass

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data (user stories, PR files, and code snippets) for processing by agents.
  • Ingestion points: SKILL.md (variables userStories, prFiles), evals/shift-left-testing.yaml (fields code_changes, code).
  • Boundary markers: None identified in the provided task definitions or agent instructions to delimit untrusted data.
  • Capability inventory: Agent-assisted test generation (qe-test-generator), requirements validation (qe-requirements-validator), and risk analysis (qe-regression-risk-analyzer) involve generating and processing code or logic.
  • Sanitization: No explicit sanitization, escaping, or validation logic for untrusted external content is defined in the skill scripts or prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 20, 2026, 07:27 AM