The Agent Skills Directory

[SAFE]: The skill provides conceptual and architectural guidance for model selection and cost optimization. No executable scripts, binaries, or dangerous command sequences were found.
[NO_CODE]: This is a documentation-only skill. It does not perform any programmatic operations or system interactions beyond providing information to the AI agent.
[INDIRECT_PROMPT_INJECTION]: The skill takes user input in the form of a task-description to provide routing advice.
Ingestion points: The task-description argument in the SKILL.md frontmatter.
Boundary markers: None explicitly defined.
Capability inventory: Limited to the Read tool; no file-writing, network communication, or command execution capabilities are present.
Sanitization: Not applicable as the skill only generates descriptive advice based on the input.

llm-router