The Agent Skills Directory

[PROMPT_INJECTION]: The file references/general/agent-skill-architecture.md contains a template titled 'Master Skill Architect System Prompt' which includes role-play instructions such as 'Role: You are an expert AI Engineer' and 'Your sole purpose is...'. These instructions could override the agent's primary behavior if it adopts the role while generating a design document.
[PROMPT_INJECTION]: The skill processes untrusted user data to produce documents, creating a surface for indirect prompt injection. 1. Ingestion points: User-provided rough notes and source materials enter the agent context via document generation tasks. 2. Boundary markers: Absent; no delimiters or 'ignore' instructions are used to wrap ingested data. 3. Capability inventory: The skill's functionality is limited to text generation and does not have access to dangerous capabilities like network operations or command execution. 4. Sanitization: Absent; the skill does not mention any validation or filtering of external content.

write-doc