The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructions in SKILL.md direct the agent to read external context files, specifically .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md, before proceeding with tasks.
Ingestion points: The agent is instructed to read local files containing product marketing context which are outside the skill's own directory but within the project environment.
Boundary markers: There are no explicit instructions to the agent to treat the content of these files as potentially untrusted or to ignore embedded instructions within them.
Capability inventory: The skill itself is instructional and does not bundle any executable scripts, binary files, or dangerous CLI tools. It relies on the agent's default capabilities for planning and documentation.
Sanitization: No sanitization or validation of the content of the marketing context files is performed.
[SAFE]: The skill provides structured templates and best practices for information architecture, including ASCII trees and Mermaid diagrams. All external links provided in the documentation, such as those to GitHub repositories for 'Verified Partners', are transparent and lead to legitimate documentation pages.

site-architecture