The Agent Skills Directory

[NO_CODE]: The skill package consists exclusively of Markdown-formatted documentation, structured templates, and validation checklists. No executable scripts (Python, JavaScript, shell) or configuration files that trigger runtime code execution were found within the provided files.
[INDIRECT_PROMPT_INJECTION]: The skill provides a framework for generating documentation from external project requirements, creating an entry point for untrusted data.
Ingestion points: The skill uses placeholders in template.md (such as {PROJECT_NAME}, {PRODUCT_NAME}, and {DESCRIPTION}) and instructions in SKILL.md to gather and format external requirement data.
Boundary markers: The templates do not implement specific delimiters or instructions for the agent to distinguish between its internal logic and potentially adversarial content within the requirements data.
Capability inventory: The skill metadata in SKILL.md enables tool access for Read, Write, and Glob operations.
Sanitization: The skill lacks explicit sanitization, validation, or escaping mechanisms for data interpolated into the document placeholders.

srs-documentation