Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Full checkout audit for the Research Project Template — tests, architecture, docs, manuscript, pipeline. USE WHEN user asks for comprehensive assessment, full repo review, health check across projects, audit everything, or pre-merge sanity sweep for template exemplars — even without naming docs/prompts or a skill. Not for single failing stage only.
.claude/skills/docxology-template-comprehensive-assessment/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | -57% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -47% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -45% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -14% | 0% |
docs/_generated/active_projects.md)docs/_generated/active_projects.md or "all active".Work through:
docs/rules/testing_standards.md).scripts/; algorithms in projects/<n>/src/ or infrastructure/. No cross-project imports._generated/active_projects.md, do not duplicate roster..bib → docs/guides/manuscript-semantics.md. Registry labels.yaml + [[FIG:]] → manuscript-cross-references. Validate markdown; PDF/logs if outputs exist.output/<name>/ per CLAUDE.md / RUN_GUIDE.bashuv sync uv run pytest projects/<project>/tests/ --cov=projects/<project>/src --cov-fail-under=90 -q uv run pytest tests/infra_tests/ --cov=infrastructure --cov-fail-under=60 -q uv run python scripts/pipeline/stage_01_test.py --project <project> uv run python -m infrastructure.validation.cli markdown projects/<project>/manuscript/ uv run python -m infrastructure.validation.cli prerender projects/<project>/manuscript --repo-root .
For a copy-paste, long-horizon launch brief that expands this workflow across software, tests, variable injection, statistics, figures, accessibility, scholarship, provenance, rendering, and release boundaries, use the comprehensive research-software and manuscript review prompt.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | pass→pass | 12,946 | 8,945 | -31% | 1 | 1 | 0% | 2,041 | 2,179 | +7% | 0 | 0 | — |
case-17 | fail→pass | 12,919 | 1,410 | -89% | 1 | 1 | 0% | 2,440 | 1,057 | -57% | 0 | 0 | — |
case-01 | fail→fail | 3,985 | 5,420 | +36% | 1 | 1 | 0% | 258 | 1,196 | +364% | 0 | 0 | — |
case-02 | fail→fail | 4,751 | 6,289 | +32% | 1 | 1 | 0% | 240 | 1,255 | +423% | 0 | 0 | — |
case-03 | fail→fail | 4,364 | 6,188 | +42% | 1 | 1 | 0% | 282 | 1,195 | +324% | 0 | 0 | — |
case-05 | pass→pass | 2,357 | 3,650 | +55% | 1 | 1 | 0% | 344 | 1,346 | +291% | 0 | 0 | — |
case-06 | pass→fail | 11,620 | 5,586 | -52% | 1 | 1 | 0% | 2,131 | 1,096 | -49% | 0 | 0 | — |
case-07 | fail→pass | 8,464 | 2,107 | -75% | 1 | 1 | 0% | 1,355 | 1,156 | -15% | 0 | 0 | — |
case-08 | fail→pass | 13,004 | 1,900 | -85% | 1 | 1 | 0% | 2,041 | 1,089 | -47% | 0 | 0 | — |
case-18 | fail→pass | 10,405 | 2,454 | -76% | 1 | 1 | 0% | 2,128 | 1,174 | -45% | 0 | 0 | — |
case-09 | fail→pass | 9,424 | 3,513 | -63% | 1 | 1 | 0% | 1,468 | 1,268 | -14% | 0 | 0 | — |
case-10 | pass→pass | 14,753 | 3,453 | -77% | 1 | 1 | 0% | 2,281 | 1,357 | -41% | 0 | 0 | — |
case-11 | pass→pass | 10,002 | 1,950 | -81% | 1 | 1 | 0% | 1,850 | 1,094 | -41% | 0 | 0 | — |
case-12 | fail→pass | 10,117 | 4,207 | -58% | 1 | 1 | 0% | 1,847 | 1,481 | -20% | 0 | 0 | — |
case-13 | pass→pass | 12,900 | 5,489 | -57% | 1 | 1 | 0% | 2,167 | 1,680 | -22% | 0 | 0 | — |
case-14 | fail→pass | 8,616 | 2,380 | -72% | 1 | 1 | 0% | 1,490 | 1,201 | -19% | 0 | 0 | — |
case-15 | fail→pass | 7,125 | 1,153 | -84% | 1 | 1 | 0% | 1,316 | 958 | -27% | 0 | 0 | — |
case-16 | pass→pass | 15,108 | 6,489 | -57% | 1 | 1 | 0% | 2,956 | 2,010 | -32% | 0 | 0 | — |
case-19 | fail→pass | 8,470 | 1,966 | -77% | 1 | 1 | 0% | 1,580 | 1,072 | -32% | 0 | 0 | — |
case-20 | fail→pass | 11,688 | 5,668 | -52% | 1 | 1 | 0% | 2,061 | 1,812 | -12% | 0 | 0 | — |
case-21 | pass→pass | 9,800 | 2,423 | -75% | 1 | 1 | 0% | 1,913 | 1,179 | -38% | 0 | 0 | — |
case-22 | pass→pass | 8,591 | 2,808 | -67% | 1 | 1 | 0% | 1,423 | 1,294 | -9% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 18 counted toward the lift figure. The other 4 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +41 percentage points is the difference between those two pass rates over the 18 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/4/2026 | +64% |
Other measured skills in the registry, with their headline benchmark lift.