qcsd-cicd-swarm

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it processes untrusted data from external sources.
  • Ingestion points: The skill reads CI/CD pipeline artifacts provided via the PIPELINE_ARTIFACTS parameter, which include test results, logs, SAST/DAST reports, and infrastructure manifests (documented in steps/01-flag-detection.md).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the artifacts are used when processing these files.
  • Capability inventory: The skill can spawn multiple sub-agents (qe-quality-gate, qe-regression-analyzer, etc.), write multiple markdown and JSON reports to the file system, and persist data across sessions using mcp__agentic-qe__memory_store (documented in steps/02-core-agents.md and steps/07-learning-persistence.md).
  • Sanitization: The instructions lack specific validation or sanitization steps for the data extracted from pipeline artifacts before it is used in decision synthesis or reporting.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 02:28 AM