sub-agent-patterns
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents patterns where sub-agents ingest data from untrusted external sources like web pages (WebFetch) or CLI outputs (npm view) and possess capabilities to modify files (Write, Edit, Bash). While the skill provides architectural advice, the specific prompt templates in SKILL.md and rules/sub-agent-patterns.md do not explicitly define boundary markers to separate untrusted data from instructions. * Ingestion points: WebFetch and Bash outputs in delegation templates. * Boundary markers: Not included in the example templates. * Capability inventory: Bash, Write, Edit, and Task (orchestration). * Sanitization: No validation or escaping of external content is suggested in the templates.
Audit Metadata