The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized actions were detected. The skill is purely instructional markdown and configuration.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it reads project-specific context from files like .agents/product-marketing.md. * Ingestion points: Reads .agents/product-marketing.md, .claude/product-marketing.md, and product-marketing-context.md. * Boundary markers: Absent; there are no instructions to ignore embedded commands in these files. * Capability inventory: No risky tools (shell, network, etc.) are utilized by this skill. * Sanitization: Absent; the content is read directly into the context.

onboarding