# Agent task validation: cloudscape

> Results cover the recorded actor profile, task suite, evidence snapshot, runtime, and run count.

Suite: `design-system-agent-tasks/1.0.0`  
Actor: `claude-code/claude-sonnet-5/cloudscape-confirmatory-controlled-v1`  
Runs: 60 / 60

Publication-ready: yes

## Observed metrics

- Task completion: 95%
- Functional task success: 43.3%
- Invalid component / prop / token uses: 7 / 38 / 0
- Correct token reuse: 83.3%
- Citation and provenance accuracy: 82.5% citation accuracy; 90% required-provenance coverage
- Citation locator validity: 89.7%
- Transparently normalized citation locators: 44
- Human interventions: 0
- Mean duration: 50.11s
- Recorded model cost: 15.537983100000002
- Model input tokens: 2111668 total (86 uncached, 1613155 cache creation, 498427 cache read)
- Model output tokens: 311251
- Exact-answer reproducibility: 36.7%
- Outcome agreement: 88.3%

Outputs can vary substantially even when functional outcomes succeed. Exact-answer reproducibility measures whether repeated runs produce the same normalized structured answer; it is not functional reliability. Different valid implementations, wording, or citation selections can lower exact-answer reproducibility even when a task completes and its runtime checks pass.

## Tasks

### foundations: cloudscape-visual-mode-token-mapping-v2

Success: 0%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.

### bindings: cloudscape-table-selection-api-binding-v2

Success: 100%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.

### components: cloudscape-controlled-toggle-preference

Success: 33.3%; exact-answer reproducibility: 33.3%; outcome agreement: 66.7%.

### structure: cloudscape-resource-inventory-empty-state-v2

Success: 66.7%; exact-answer reproducibility: 33.3%; outcome agreement: 66.7%.

### structure: cloudscape-create-configure-save-flow

Success: 0%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.

### governance: cloudscape-release-evidence-governance-v2

Success: 0%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.

### foundations: cloudscape-focus-visible-semantic-pair-v1

Success: 100%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.

### foundations: cloudscape-responsive-content-spacing-v1

Success: 100%; exact-answer reproducibility: 66.7%; outcome agreement: 100%.

### foundations: cloudscape-dark-surface-token-boundary-v1

Success: 0%; exact-answer reproducibility: 66.7%; outcome agreement: 100%.

### bindings: cloudscape-modal-api-binding-v2

Success: 66.7%; exact-answer reproducibility: 33.3%; outcome agreement: 66.7%.

### components: cloudscape-flashbar-recovery-message-v2

Success: 0%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.

### governance: cloudscape-accessibility-regression-governance-v2

Success: 66.7%; exact-answer reproducibility: 33.3%; outcome agreement: 66.7%.

### bindings: cloudscape-navigation-follow-binding-v3

Success: 0%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.

### bindings: cloudscape-required-name-input-binding-v2

Success: 0%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.

### components: cloudscape-modal-open-close-behavior-v3

Success: 33.3%; exact-answer reproducibility: 33.3%; outcome agreement: 66.7%.

### structure: cloudscape-filtered-inventory-empty-state-v1

Success: 0%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.

### governance: cloudscape-accessibility-fix-review-governance-v1

Success: 66.7%; exact-answer reproducibility: 33.3%; outcome agreement: 66.7%.

### components: cloudscape-button-pending-label-behavior-v2

Success: 100%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.

### structure: cloudscape-load-more-success-flow-v4

Success: 33.3%; exact-answer reproducibility: 33.3%; outcome agreement: 66.7%.

### governance: cloudscape-public-change-review-governance-v3

Success: 100%; exact-answer reproducibility: 33.3%; outcome agreement: 100%.