The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill acts as a documentation and template repository for Mermaid diagramming.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the transformation of user-supplied requirements into Mermaid diagram code. While this creates a theoretical surface for indirect prompt injection via malicious data content, the skill proactively mitigates this by providing extensive security documentation and safety guidance.
Ingestion points: Processes user requirements and data arrays to generate diagram DSL code.
Boundary markers: Includes validation checklists in SKILL.md and references/validation.md to ensure proper syntax and structure.
Capability inventory: The generated code supports interactive features like clickable links (href) and JavaScript callbacks.
Sanitization: The skill provides clear instructions on configuring Mermaid's securityLevel (strict, antiscript, sandbox) and using DOMPurify to sanitize HTML label content, specifically warning against javascript: and data: URLs.
[EXTERNAL_DOWNLOADS]: The skill references standard external assets and documentation from trusted or well-known sources.
Icons: Fetches icons from the well-known iconify.design service.
Documentation: Example templates include links to official documentation for reputable services including Stripe, Elasticsearch, Redis, PostgreSQL, MongoDB, and Apache Kafka.
[OBFUSCATION]: The skill documents the use of HTML entities (e.g., # for #) within Mermaid labels. This is a functional requirement of the Mermaid DSL for character escaping and is not used for concealment or malicious purposes.

mermaid-diagramming