Evidence Insight

cellxgene-census

86100Total Score

Core Capability

84 / 100

Functional Suitability

11 / 12

Reliability

9 / 12

Performance & Context

7 / 8

Agent Usability

14 / 16

Human Usability

8 / 8

Security

10 / 12

Maintainability

9 / 12

Agent-Specific

16 / 20

Medical Task

20 / 20 Passed

92Cross-tissue or cross-disease expression comparisons (e.g., macrophages across lung/liver/brain; COVID-19 vs control)

4/4

88Reference atlas lookups to contextualize findings from your own single-cell dataset (marker validation, expected expression patterns)

4/4

86Programmatic access to versioned CZ CELLxGENE Census data (human and mouse)

4/4

86Query cell (obs) metadata and gene (var) metadata with expressive filter syntax

4/4

86End-to-end case for Programmatic access to versioned CZ CELLxGENE Census data (human and mouse)

4/4

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

PASS

Research Veto✅ PASS — Applicable

Dimension	Result	Detail
Scientific Integrity	PASS	The legacy audit did not indicate that retrieval outputs were presented as unsupported findings.
Practice Boundaries	PASS	Practice boundaries held because the package remained focused on source handling, lookup, or structured evidence use.
Methodological Ground	PASS	The legacy audit preserved a method-grounded interpretation of the Programmatically query the CZ CELLxGENE Census (61M+ cells) when you need cross-tissue, disease, or cell-type expression data for population-scale queries and reference atlas comparisons workflow.
Code Usability	N/A	This package is packaging-first and output-first, not code-first, so code usability is treated as not applicable.

Core Capability84 / 100 — 8 Categories

Functional Suitability

The archived deduction in functional suitability traces back to: Improve stress-case output rigor. Stress and boundary scenarios show weaker consistency

11 / 12

92%

Reliability

The archived deduction in reliability traces back to: Improve stress-case output rigor. Stress and boundary scenarios show weaker consistency

9 / 12

75%

Performance & Context

The legacy audit deducted points for cellxgene-census in performance context.

7 / 8

88%

Agent Usability

A modest deduction remained in agent usability for cellxgene-census in the archived review.

14 / 16

88%

Human Usability

Human usability reached full score in the archived evaluation.

8 / 8

100%

Security

The archived evaluation left some headroom for cellxgene-census under security.

10 / 12

83%

Maintainability

The archived evaluation left some headroom for cellxgene-census under maintainability.

9 / 12

75%

Agent-Specific

Related legacy finding for cellxgene-census: Improve stress-case output rigor. Stress and boundary scenarios show weaker consistency

16 / 20

80%

Core Capability Total84 / 100

Medical TaskExecution Average: 87.6 / 100 — Assertions: 20/20 Passed

Canonical

Cross-tissue or cross-disease expression comparisons (e.g., macrophages across lung/liver/brain; COVID-19 vs control)

4/4 ✓

Variant A

Reference atlas lookups to contextualize findings from your own single-cell dataset (marker validation, expected expression patterns)

4/4 ✓

Edge

Programmatic access to versioned CZ CELLxGENE Census data (human and mouse)

4/4 ✓

Variant B

Query cell (obs) metadata and gene (var) metadata with expressive filter syntax

4/4 ✓

Stress

End-to-end case for Programmatic access to versioned CZ CELLxGENE Census data (human and mouse)

4/4 ✓

Canonical✅ Pass

Cross-tissue or cross-disease expression comparisons (e.g., macrophages across lung/liver/brain; COVID-19 vs control)

Cross-tissue or cross-disease expression comparisons (e.g.,... was evaluated as a bounded documentation path, not as a runnable script workflow.

Basic 36/40|Specialized 56/60|Total 92/100

✅A1The cellxgene-census output structure matches the documented deliverable

✅A2The instruction path remains actionable for the documented case

✅A3The output stays fully within the documented skill boundary

✅A4The response quality is acceptable for the documented path

Pass rate: 4 / 4

Variant A✅ Pass

Reference atlas lookups to contextualize findings from your own single-cell dataset (marker validation, expected expression patterns)

The archived run for Reference atlas lookups to contextualize findings from your own... remained guidance-driven rather than command-driven.

Basic 34/40|Specialized 54/60|Total 88/100

✅A1The cellxgene-census output structure matches the documented deliverable

✅A2The instruction path remains actionable for the documented case

✅A3The output stays fully within the documented skill boundary

✅A4The response quality is acceptable for the documented path

Pass rate: 4 / 4

Edge✅ Pass

Programmatic access to versioned CZ CELLxGENE Census data (human and mouse)

Programmatic access to versioned CZ CELLxGENE Census data (human... was evaluated as a bounded documentation path, not as a runnable script workflow.

Basic 33/40|Specialized 53/60|Total 86/100

✅A1The cellxgene-census output structure matches the documented deliverable

✅A2The instruction path remains actionable for the documented case

✅A3The output stays fully within the documented skill boundary

✅A4The response quality is acceptable for the documented path

Pass rate: 4 / 4

Variant B✅ Pass

Query cell (obs) metadata and gene (var) metadata with expressive filter syntax

The archived run for Query cell (obs) metadata and gene (var) metadata with expressive... remained guidance-driven rather than command-driven.

Basic 32/40|Specialized 54/60|Total 86/100

✅A1The cellxgene-census output structure matches the documented deliverable

✅A2The instruction path remains actionable for the documented case

✅A3The output stays fully within the documented skill boundary

✅A4The response quality is acceptable for the documented path

Pass rate: 4 / 4

Stress✅ Pass

End-to-end case for Programmatic access to versioned CZ CELLxGENE Census data (human and mouse)

End-to-end case for Programmatic access to versioned CZ CELLxGENE... was evaluated as a bounded documentation path, not as a runnable script workflow.

Basic 29/40|Specialized 57/60|Total 86/100

✅A1The cellxgene-census output structure matches the documented deliverable

✅A2The instruction path remains actionable for the documented case

✅A3The output stays fully within the documented skill boundary

✅A4The response quality is acceptable for the documented path

Pass rate: 4 / 4

Medical Task Total87.6 / 100

Key Strengths

Primary routing is Evidence Insight with execution mode A
Static quality score is 84/100 and dynamic average is 79.6/100
Assertions and command execution outcomes are recorded per input for human review
Execution verification summary: No script verification was applicable