The Agent Skills Directory

[SAFE]: No evidence of prompt injection or malicious instructions was found. The skill operates as a structured documentation generator based on user requirements.
[SAFE]: The skill does not contain hardcoded secrets or sensitive data. Templates correctly use environment variable patterns (e.g., process.env.[KEY_NAME]) and clearly labeled placeholders (e.g., sk_test_...) for authentication configuration.
[SAFE]: External references are limited to well-known and trusted services, including Cloudflare (Workers, D1, R2), Clerk, Stripe, and OpenAI. All referenced tools and packages are standard in the modern web development ecosystem.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest user requirements and output documentation files, representing an attack surface for indirect prompt injection.
Ingestion points: User project descriptions and functional requirements processed by the project-planning workflow.
Boundary markers: None explicitly defined for untrusted input handling.
Capability inventory: Writing multiple Markdown documents to the local workspace.
Sanitization: None explicitly implemented; the skill populates templates with raw user input. However, this is expected behavior for a developer planning tool and carries minimal risk.

project-planning