The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied information regarding products and target markets to fill a predefined template. Although this represents an ingestion point for untrusted data, the skill possesses no capabilities—such as file manipulation, network requests, or command execution—that could be leveraged for an attack.
Ingestion points: SKILL.md Input section (user-provided product and customer details).
Boundary markers: Absent.
Capability inventory: No tools or executable actions are defined or requested.
Sanitization: Absent.
[NO_CODE]: The provided skill files (SKILL.md, template.md, and examples) consist exclusively of Markdown text, instructions, and static templates. There are no scripts, environment configurations, or dependencies that execute on the host system or within the agent's environment.

positioning-statement