modelchk

1Size the run before spending it: how strong a model, and how hard it should think.
2 
3## Goal
4 
5`modelchk` is read-only and advisory. One assessment of risk and complexity sizes two dials: capability tier and reasoning effort. It does not choose, route, switch, pin, spawn, set, or require any concrete model or level.
6 
7**Capability tier** is the cheapest sufficient class:
8 
9- `fast` for local, mechanical, reversible work with cheap, complete verification.
10- `standard` for ordinary repo-grounded reasoning, multi-step drafting, normal coding, and conventional documentation or skill work.
11- `frontier` for architecture, high ambiguity, safety/security/privacy/data-loss risk, release-critical review, cross-domain scope, or work where one wrong assumption wastes a large run.
12 
13**Reasoning effort** is how hard the model should deliberate. The skill recommends an intent; the executor chooses the model and resolves that intent to its actual level. From least to most deliberation:
14 
15- `glance`: minimal deliberation; take the direct path. Resolves to the model's floor.
16- `measured`: ordinary deliberation. Resolves to the model's default, or the middle of its ladder if no default is named.
17- `thorough`: work the alternatives and check assumptions. Resolves above the everyday setting, short of the top.
18- `exhaustive`: maximal deliberation; exhaust the search and re-check the work. Resolves to the model's ceiling.
19…