skills/sickn33/agentic-awesome-skills/full-stack-orchestration-full-stack-feature/Gen Agent Trust Hub
full-stack-orchestration-full-stack-feature
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface where untrusted user input is interpolated into prompts for specialized subagents without adequate safety boundaries.
- Ingestion points: The
$ARGUMENTSplaceholder, representing user-provided feature requirements, is interpolated into twelve different prompts across all phases inSKILL.md. - Boundary markers: The prompts do not use delimiters (e.g., triple quotes, XML tags) or explicit instructions to the subagents to ignore instructions embedded within the
$ARGUMENTStext. - Capability inventory: The orchestration involves high-privilege operations performed by subagents, including database schema modification, code implementation, infrastructure setup (Terraform/Kubernetes), and CI/CD pipeline configuration.
- Sanitization: The instructions lack any input validation, sanitization, or filtering steps before passing the user data to the
Tasktool subagents.
Audit Metadata