The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the skill instructions or configuration.
The skill is purely instructional, using Markdown to guide the agent's behavior without the use of scripts (.py, .js, .sh) or binaries.
The platform configuration explicitly disables model invocation, and no tools or network permissions are requested.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-supplied plans, creating a standard ingestion surface for untrusted data.
Ingestion points: User-provided 'plan, decision, or idea' passed as an argument (SKILL.md).
Boundary markers: None present in the prompt template.
Capability inventory: The skill has no capabilities to execute commands, write files, or access the network, rendering the injection surface benign.
Sanitization: Not implemented, but unnecessary given the lack of executable actions.

grill-me