# Benchmark Coverage Gap Map (6 Charter Capability Classes)

## Scope and evidence
- Source manifest: `/mnt/c/_organized_non_vital/ai_research/benchmark_manifest.json`
- This map file: `ai_research/benchmarks/COVERAGE_GAP_MAP.md`
- Audit script: `ai_research/benchmarks/coverage_gap_audit.py`
- Snapshot timestamp: 2026-03-04T13:13:50.484852Z

## Required charter classes
1. Perception & Grounding
2. Reasoning & Inference
3. Planning & Control
4. Memory & State Tracking
5. Tool Use & Environment Interaction
6. Reliability, Safety & Maintainability

## Explicit mapping from `benchmark_manifest.json` entries

| benchmark_id | benchmark_name | mapped_class_id | mapped_class_name | command |
|---|---|---:|---|---|

## Class-level coverage map

| class_id | class_name | coverage_status | mapped_benchmark_ids | interpretation |
|---|---|---|---|---|
| 1 | Perception & Grounding | uncovered | `[]` | No dedicated benchmark currently validates this charter class. |
| 2 | Reasoning & Inference | uncovered | `[]` | No dedicated benchmark currently validates this charter class. |
| 3 | Planning & Control | uncovered | `[]` | No dedicated benchmark currently validates this charter class. |
| 4 | Memory & State Tracking | uncovered | `[]` | No dedicated benchmark currently validates this charter class. |
| 5 | Tool Use & Environment Interaction | uncovered | `[]` | No dedicated benchmark currently validates this charter class. |
| 6 | Reliability, Safety & Maintainability | uncovered | `[]` | No dedicated benchmark currently validates this charter class. |

## Uncovered class IDs
Explicit uncovered class IDs: [1, 2, 3, 4, 5, 6]

## Shell-executable proposals for uncovered classes
### Shell-Executable Proposal for Class 1
```bash
# TODO: add benchmark coverage command for class 1.
python3 ai_research/benchmarks/coverage_gap_audit.py --manifest ai_research/benchmark_manifest.json
```

### Shell-Executable Proposal for Class 2
```bash
# TODO: add benchmark coverage command for class 2.
python3 ai_research/benchmarks/coverage_gap_audit.py --manifest ai_research/benchmark_manifest.json
```

### Shell-Executable Proposal for Class 3
```bash
# TODO: add benchmark coverage command for class 3.
python3 ai_research/benchmarks/coverage_gap_audit.py --manifest ai_research/benchmark_manifest.json
```

### Shell-Executable Proposal for Class 4
```bash
# TODO: add benchmark coverage command for class 4.
python3 ai_research/benchmarks/coverage_gap_audit.py --manifest ai_research/benchmark_manifest.json
```

### Shell-Executable Proposal for Class 5
```bash
# TODO: add benchmark coverage command for class 5.
python3 ai_research/benchmarks/coverage_gap_audit.py --manifest ai_research/benchmark_manifest.json
```

### Shell-Executable Proposal for Class 6
```bash
# TODO: add benchmark coverage command for class 6.
python3 ai_research/benchmarks/coverage_gap_audit.py --manifest ai_research/benchmark_manifest.json
```

## Shell-checkable summary (reproducible)

Run:
```bash
python3 ai_research/benchmarks/coverage_gap_audit.py --manifest ai_research/benchmark_manifest.json
```

- `total_required_classes: 6`
- `covered_class_count: 0`
- `uncovered_class_count: 6`
- `covered_ratio: 0/6`
- `uncovered_class_ids: [1, 2, 3, 4, 5, 6]`

Script source:
- `python3 ai_research/benchmarks/coverage_gap_audit.py`

## Decision quality note
Only class 6 has benchmark coverage in current manifest; classes 1-5 are coverage gaps and need new benchmark entries.