The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and iterate on untrusted user input, such as project ideas and requirements, to generate design documentation.
Ingestion points: User-supplied descriptions of ideas, goals, and constraints provided during the dialogue phases in SKILL.md.
Boundary markers: No specific delimiters or instructions to ignore embedded commands within user input are defined in the skill instructions.
Capability inventory: The skill utilizes the AskUserQuestion tool for structured interaction. It does not perform file writes, network requests, or command execution within its own logic.
Sanitization: There is no evidence of filtering or sanitization of user-provided content before it is incorporated into the design process.
Context: The risk is minimal as the skill's purpose is conceptual design and documentation rather than automated system modification or code execution.
[NO_CODE]: The skill consists exclusively of Markdown instructions (SKILL.md), reference examples, and documentation templates (assets/). It does not include scripts, binaries, or instructions to install third-party dependencies.
[SAFE]: No patterns of malicious intent, such as prompt injection, obfuscation, data exfiltration, or persistence mechanisms, were found. The skill adheres to the stated purpose of assisting with software design workflows.

brainstorming