The Agent Skills Directory

[SAFE]: Analysis of the skill instructions and reference materials revealed no malicious commands, obfuscated content, or unauthorized network operations.- [PROMPT_INJECTION]: The skill is configured to ingest project context from local markdown files, specifically .agents/product-marketing-context.md. This represents a surface for indirect prompt injection if these local files are modified by an untrusted source, although the skill lacks the tools necessary to perform high-risk actions.
Ingestion points: SKILL.md (references .agents/product-marketing-context.md and .claude/product-marketing-context.md)
Boundary markers: Absent in instructions
Capability inventory: None (no tools requested in frontmatter)
Sanitization: Absent

onboarding-cro