The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted user requirements to generate document artifacts, creating a surface for indirect prompt injection.
Ingestion points: Untrusted data is gathered via AskUserQuestion and user prompts for sections like problem statements and evidence in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the generated PRD sections.
Capability inventory: The skill uses Read and Write tools in SKILL.md and executes a local Node.js script scripts/main.cjs.
Sanitization: No input validation or sanitization logic is described in the workflow or scripts.

prd-generator