The Agent Skills Directory

[DATA_EXPOSURE]: The skill recommends reading project context files (such as .claude/project-context.md or .cursor/project-context.md) to determine site hierarchy. This is a standard practice for development agents to align with the local project structure and does not involve accessing sensitive system credentials or private data.
[INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface as it processes local project files to identify page types and site depth. However, this is the primary purpose of the skill to generate accurate breadcrumbs. The risk is mitigated as the skill only generates static HTML and JSON-LD structured data and does not have capabilities for command execution or remote network operations based on the ingested content.
[EXTERNAL_DOWNLOADS]: The skill references external validation services including the Google Rich Results Test and the Schema Markup Validator. These are official, well-known industry tools used for verifying structured data and are documented here as trusted resources.

breadcrumb-generator