/cheese

Use this skill as the single front door to the easy-cheese workflow. Inspect whatever the user dropped in, classify it into an intent shape, announce the routing decision as a short three-line block (Intent / Reason / Target — see ## Output), and dispatch immediately to the chosen skill. Cheese is autonomous by default — it picks the best target and runs it, only stopping to ask when --safe is passed or when the input is genuinely ambiguous.

Do not use it once a downstream skill is already running, or when the user has already named the skill they want (/mold ..., /cook ..., /age, etc.) — pass straight through to that skill instead.

Inputs

Accept anything the user supplies as $ARGUMENTS:

• A natural-language feature description, idea, or question.
• A spec path (.cheese/specs/<slug>.md) or pasted spec content.
• A bug report, stack trace, failing test output, or reproduction steps.
• A file path, glob, or directory.
• A PR or issue reference (PR#142, #87, GitHub URL).
• A research question about an external library, API, or pattern.
• An empty or near-empty prompt — treat as "what's next?" and clarify.

Optional flags: