Skill
Skill answers one question: how well are you prompting? The LLM judge looks at seven habits on every scored prompt and marks each as present or missing. The Skill page rolls those habits up into a single Skill Index.
Headline — Skill Index
Section titled “Headline — Skill Index”Skill Index (0–100). A habit-based score computed by the API from the seven rubric checks across the window. It maps to a Skill tier — the score hub has a Completion tier with the same five names at different cuts, so read the label beside the ladder:
| Skill tier | Skill Index |
|---|---|
| Novice | below 20 |
| Practitioner | 20–35 |
| Proficient | 35–50 |
| Expert | 50–65 |
| Elite | 65+ |
Skill is the habit view. For the per-prompt letter grades behind it, see Prompt Grade.
The rest of the page
Section titled “The rest of the page”| Element | What it tells you |
|---|---|
| Skill Index over time | The index per period, drawn over the shaded tier bands so you can see which tier each period sat in. |
| Per-pillar breakdown | The share of prompts where the judge marked each habit as present, over the last 30 days. The weakest pillar is badged — that is what to work on first. |
The four pillars
Section titled “The four pillars”The seven rubric checks group into four pillars:
| Pillar | Question it asks |
|---|---|
| Clarity | Did you tell the model what you actually want? |
| Context | Did you give the model what it needs to answer? |
| Verification | Did you ask the model to check its work? |
| Workflow | Did you keep the session tight and on track? |
The seven checks
Section titled “The seven checks”| Check | Hint |
|---|---|
| Goal explicit | States the outcome in concrete, observable terms |
| Scope bounded | Limits files, features, or surface area touched |
| Concrete references | Cites files, lines, or @-references the model can open |
| Enough context | Includes errors, examples, or artifacts the model needs |
| Verification requested | Asks for tests, expected output, or success criteria |
| Root-cause oriented | Targets the underlying cause, not just the symptom |
| Plan first | Opens non-trivial work with an explore-or-plan turn before code |
Not every check applies to every prompt — the judge marks each as applicable based on intent (fix, plan, explore, etc.).
What is excluded
Section titled “What is excluded”Prompts the judge couldn’t reach a clean verdict on are left out entirely, so one bad grading run doesn’t drag your score around.
The Skill Index headline and trend are coloured by the five tier bands above (Novice → Elite, at 20 / 35 / 50 / 65). The index has a natural ceiling around 60–70, because some habits don’t apply to every prompt, so Elite is deliberately rare.
The per-pillar rubric bars use their own four-band colour scale: Poor (0–30) · Weak (30–50) · Fair (50–70) · Good (70–100).
For the per-prompt letter grade (A+ ≥ 90 … F below 12) and its bands, see the Prompt Grade page.
Why it matters
Section titled “Why it matters”Speed measures throughput; Skill measures how clean the inputs were that produced it. High Speed with low Skill usually means you are getting lucky — improving Skill is what makes the throughput repeatable.