skill-model-gap — Claude Code skill: what it does, how to install
skill-model-gap is a Claude Code skill that pinpoints why an agent instruction performs worse on a target model than on a reference model and fixes the prompt text.
MIT ★ 262 updated 2026-11-06
/plugin marketplace add nicobailon/skill-diet
/plugin install skill-diet@skill-dietWhat skill-model-gap does
The skill evaluates task outcomes between a high-performing reference model and an underperforming target model. It executes tests using the skill-diet harness via node ../skill-diet/scripts/eval.mjs. Tests run on a single host and adapter, isolating model variance from environment differences.
Evaluation inspects instruction loading before prompt rules: when a target model skips reading SKILL.md, the fix adjusts description triggers. Reviewers label log failures, and each isolated edit is tested as an L<n> variant over multiple passes. Running the workflow requires the skill-diet evaluation suite, Node.js, and an active API token budget to cover repeated benchmark runs.
What it helps with
- diagnose skill regressions between different model targets
- verify whether a model consistently opens SKILL.md before execution
- tune prompt rules one change at a time without breaking baseline models
- measure token spending across repeated benchmark passes
Example prompts for the agent
- Compare this skill across Claude 3.5 Sonnet and GPT-4o using skill-model-gap
- Find why the target model fails skill tasks that pass on reference
How to install skill-model-gap
1Plugin marketplace
/plugin marketplace add nicobailon/skill-diet/plugin install skill-diet@skill-dietThis installs the whole skill-diet plugin with all its skills
2Skills CLI (Vercel)
npx skills add nicobailon/skill-diet --skill skill-model-gap -a claude-code3Manually: put the files in place
~/.claude/skills/skill-model-gap/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/skill-model-gap/
1Skills CLI (Vercel)
npx skills add nicobailon/skill-diet --skill skill-model-gap -a codex2Manually: put the files in place
.agents/skills/skill-model-gap/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add nicobailon/skill-diet --skill skill-model-gap -a cursor2Manually: put the files in place
.cursor/skills/skill-model-gap/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install nicobailon/skill-diet skills/skill-model-gap2Manually: put the files in place
.github/skills/skill-model-gap/SKILL.md1Skills CLI (Vercel)
npx skills add nicobailon/skill-diet --skill skill-model-gap -a gemini-cli2Platform CLI
gemini skills install https://github.com/nicobailon/skill-diet --path skills/skill-model-gapManually: put the files in place
.devin/skills/skill-model-gap/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about skill-model-gap
How to install skill-model-gap in Claude Code?
Run both marketplace commands in order: /plugin marketplace add nicobailon/skill-diet followed by /plugin install skill-diet@skill-diet.
What does skill-model-gap do?
It calculates accuracy gaps across models executing the same skill, diagnoses failure logs, and tests single-change fixes against held-back tasks.
What is required to run skill-model-gap?
It requires the skill-diet harness scripts, Node.js, and paid API quota for every model included in the test runs.
Source text
SKILL.md · MIT
--- name: skill-model-gap description: Find why a skill works worse on one model than another, and fix it without hurting the others. Give it the skill and the models to compare. disable-model-invocation: true metadata: version: 0.2.0 creator: Nico Bailon --- # Skill model gap A skill tested on one model (the **reference**) often passes fewer tasks on another (the **target**). Find where and why, fix it with text every model reads, and prove the target gains while no model drops. This runs skill-diet's eval: tasks and scoring per [eval.md](../skill-diet/references/eval.md), the adapter per [adapter.md](../skill-diet/references/adapter.md), commands as `node ../skill-diet/scripts/eval.mjs …`. Edits follow skill-diet's [Editing a skill](../skill-diet/SKILL.md#editing-a-skill). ## Set up - **One adapter, one host** for every model, so only the model differs. Pass `--model` to every `run` and `score`. - **Tasks** per eval.md, each labeled with the case it covers. Split them blind and at random: about a third, at least one per case, go to a **held-back** folder whose `adapter` and `sk/` link to the tuning folder's. Held-back tasks get no runs until the end. - **Versions:** `base` (the skill as it is), `none`, and one folder per candidate fix. - **Models:** the reference has the best `base` pass rate (on a tie, the one that reads the skill most); every other model with a gap is a target. Add at least one more model as a harm check. - **Paid quota:** before each step, give the user sessions × cost per run and stay within the budget they set. Two tasks first measure the cost. ## 1. Measure the gap `base` and `none` on every model, tuning tasks, 3 runs. `score <dir> --models <all> --versions none,base` puts the models side by side; `--model <m>` instead gives one model's warnings. A gap is a task the target fails more often than the reference with a cause in its logs; 2/3 against 3/3 alone is noise. If the target does better with `none` (MISS or WORSE here), look for lines to cut first. No gap on any target → stop and report. ## 2. Find the cause **Loading first.** Compare how many runs read `SKILL.md` on each model (the `--models` table's last row; `score`'s under-half warning is this finding). If the target opens the skill much less often, fix loading before any rule: a description that says when to read the skill and what it adds to the tool's own description. If reads rise but passes don't, move the rules the target breaks into the description. Otherwise a fresh reviewer reads each failing target run beside a passing reference run of the same task, and labels one cause with quoted evidence: - skipped, misread or found no rule for the case in the skill it loaded → reword or widen the rule, or move it up a level; - stopped early or asked the user → one line on what done means, unless the task really needs a question; - tool, fixture or adapter (wrong parameter, tool error, missing file, dropped output) → fix it there, not in the skill; - can't do it (fails even with the reference's steps written into the prompt) → drop it from the target's goal and report it; - valid but scored wrong → eval.md's scorer check; fix the task and rerun it in every version on every model. Each cause needs a held-back task that hits it (the same command or call). Write any that are missing now, before any fix, so step 4 can show a gain. ## 3. Fix one change at a time Try a shared fix first: plainer, positive, more direct wording helps every model. Each candidate is a version `L<n>`: `base` plus one change. 1. **Helps the target:** `base` vs `L<n>` on the target, the gap tasks it's for, 6 runs. Keep it if the gain `score <dir> --model <target> --versions base,L<n> --tasks <those tasks>` prints is larger than its 2 SE, or a task flips from failing every run to passing every run. 2. **Hurts no one:** all kept changes as `tuned`, vs `base` on every other model, all tuning tasks, 3 runs. On WORSE or MISS, leave one change out at a time to find it, then drop or reword it. Compare reads too: a model that stops opening the skill loses the rules left in its body, so report the drop. 3. **Still helps together:** `tuned` vs `base` on the target, all tuning tasks. Every model reads every line; a host can't reliably hide a line by model. ## 4. Held-back, once `base` vs `tuned` on every model, held-back folder, 9 runs. Ship when the target gains more than 2 SE, no model is WORSE or MISSes, and `SKILL.md` stays within skill-diet's budget. Report tokens/run per model as a cost: a skill that models now open costs its size on every task. If `base` already passes the target's held-back tasks, they can't show a gain: add tasks for the gap's causes and run once more. If `base` fails them and `tuned` doesn't gain, the fixes are overfit; write new held-back tasks before tuning again. ## Afterward On ship, record every model as eval.md's [Tested with](../skill-diet/references/eval.md#tested-with) says. The commit message names which change came from which target, with the measurements. When a target in `tested-with` is replaced, rerun `base` with and without its changes on the new model, and delete what no longer helps.
© Nico Bailon · MIT · nicobailon/skill-diet