Data Analysis
gene-protein-expression-matrix-normalization
Normalize gene or protein expression matrices using log2, z-score, quantile normalization, or TMM. Inputs: raw expression matrix. Outputs: normalized matrix, density distribution plots, before-after boxplots, QC summary report.
93100Total Score
Core Capability
93 / 100
Functional Suitability
12 / 12
Reliability
11 / 12
Performance & Context
7 / 8
Agent Usability
15 / 16
Human Usability
8 / 8
Security
11 / 12
Maintainability
11 / 12
Agent-Specific
18 / 20
Medical Task
25 / 25 Passed
94log2 normalization smoke test
5/5
93zscore normalization smoke test
5/5
93minmax normalization smoke test
5/5
91CLI help and parameter contract
5/5
90bundled output reproducibility review
5/5
Veto GatesRequired pass for any deployment consideration
Skill Veto✓ All 4 gates passed
✓
Operational Stability
System remains stable across varied inputs and edge cases
PASS✓
Structural Consistency
Output structure conforms to expected skill contract format
PASS✓
Result Determinism
Equivalent inputs produce semantically equivalent outputs
PASS✓
System Security
No prompt injection, data leakage, or unsafe tool use detected
PASSResearch Veto✅ PASS — Applicable
| Dimension | Result | Detail |
|---|---|---|
| Scientific Integrity | PASS | No fabricated DOI, PMID, trial result, sample size, p-value, or unsupported scientific claim was generated during audit. |
| Practice Boundaries | PASS | The skill performs computational data analysis and does not make diagnostic or treatment recommendations. |
| Methodological Ground | PASS | The workflow uses standard data-analysis methods and documents assumptions, thresholds, and output interpretation boundaries. |
| Code Usability | PASS | Native CLI execution was verified using /opt/homebrew/bin/Rscript in this environment. |
Core Capability93 / 100 — 8 Categories
Functional Suitability
The skill provides a complete Data Analysis workflow with documented inputs, outputs, examples, and deterministic artifacts.
12 / 12
100%
Reliability
Native execution was verified with /opt/homebrew/bin/Rscript; remaining risks are limited to environment dependency drift.
11 / 12
92%
Performance & Context
Runtime is appropriate for bundled smoke-test data and the workflow writes concise tabular and plot artifacts.
7 / 8
88%
Agent Usability
CLI usage, parameters, output paths, and troubleshooting guidance are sufficiently clear for agent invocation.
15 / 16
94%
Human Usability
Examples and reference documentation make the workflow discoverable and reproducible for human reviewers.
8 / 8
100%
Security
No credential handling or unsafe dynamic code execution was identified; file-path based inputs are used.
11 / 12
92%
Maintainability
Implementation is modular across scripts, references, and tests, making future updates straightforward.
11 / 12
92%
Agent-Specific
The skill has clear trigger boundaries, structured CLI execution, and reproducible output conventions.
18 / 20
90%
Core Capability Total93 / 100
Medical TaskExecution Average: 92.2 / 100 — Assertions: 25/25 Passed
94
Canonical
log2 normalization smoke test
5/5 ✓
93
Variant A
zscore normalization smoke test
5/5 ✓
93
Variant B
minmax normalization smoke test
5/5 ✓
91
Edge
CLI help and parameter contract
5/5 ✓
90
Stress
bundled output reproducibility review
5/5 ✓
94
Canonical✅ Pass
log2 normalization smoke test
/opt/homebrew/bin/Rscript completed log2 normalization and wrote normalized_matrix, summaries, manifest, and session info.
Basic 38/40|Specialized 56/60|Total 94/100
✅A1Required output artifacts were generated or documented for this test input.
✅A2Input validation and documented parameter handling were consistent with the skill scope.
✅A3No fabricated biomedical claims or unsupported clinical conclusions were generated.
✅A4Execution stayed within the Data Analysis workflow boundaries.
✅A5Results were reproducible enough for audit review.
Pass rate: 5 / 5
93
Variant A✅ Pass
zscore normalization smoke test
The z-score route completed successfully on bundled expression_matrix.csv and produced expected outputs.
Basic 37/40|Specialized 56/60|Total 93/100
✅A1Required output artifacts were generated or documented for this test input.
✅A2Input validation and documented parameter handling were consistent with the skill scope.
✅A3No fabricated biomedical claims or unsupported clinical conclusions were generated.
✅A4Execution stayed within the Data Analysis workflow boundaries.
✅A5Results were reproducible enough for audit review.
Pass rate: 5 / 5
93
Variant B✅ Pass
minmax normalization smoke test
The min-max route completed successfully and wrote table and RDS artifacts.
Basic 37/40|Specialized 56/60|Total 93/100
✅A1Required output artifacts were generated or documented for this test input.
✅A2Input validation and documented parameter handling were consistent with the skill scope.
✅A3No fabricated biomedical claims or unsupported clinical conclusions were generated.
✅A4Execution stayed within the Data Analysis workflow boundaries.
✅A5Results were reproducible enough for audit review.
Pass rate: 5 / 5
91
Edge✅ Pass
CLI help and parameter contract
The documented CLI help rendered successfully with method, margin, pseudo-count, centering, scaling, delimiter, and seed options.
Basic 36/40|Specialized 55/60|Total 91/100
✅A1Required output artifacts were generated or documented for this test input.
✅A2Input validation and documented parameter handling were consistent with the skill scope.
✅A3No fabricated biomedical claims or unsupported clinical conclusions were generated.
✅A4Execution stayed within the Data Analysis workflow boundaries.
✅A5Results were reproducible enough for audit review.
Pass rate: 5 / 5
90
Stress✅ Pass
bundled output reproducibility review
Pre-generated outputs for log2, zscore, and minmax provide reproducibility evidence and expected artifact structure.
Basic 36/40|Specialized 54/60|Total 90/100
✅A1Required output artifacts were generated or documented for this test input.
✅A2Input validation and documented parameter handling were consistent with the skill scope.
✅A3No fabricated biomedical claims or unsupported clinical conclusions were generated.
✅A4Execution stayed within the Data Analysis workflow boundaries.
✅A5Results were reproducible enough for audit review.
Pass rate: 5 / 5
Medical Task Total92.2 / 100
Key Strengths
- Native R execution succeeded for log2, zscore, and minmax normalization modes.
- The skill has deterministic transformations with clear output tables and RDS artifacts.
- Documentation and scripts are modular, with algorithm, CLI, and troubleshooting references.
- Bundled outputs make frontend and reviewer inspection straightforward.