---
name: undominated-tool-selection-recall-audit
description: "Audits tool routing accuracy across large tool inventories and detects hallucinated function invocations. Recomputed fields must match the other inputs. A pass is not a compliance, safety, or price verdict."
license: MIT
metadata:
  author: Undominated.ai
  version: "1.1.0"
---

# Tool Selection Recall Audit

## Description
Audits tool routing accuracy across large tool inventories and detects hallucinated function invocations.

## Category
`tool-use`

## Tags
`tool-selection`, `function-calling`, `recall`, `hallucination`

## When to Use
Benchmark tool routing pipelines when providing models with dozens of available tool definitions.

## Operational Limitations
Evaluates precision and recall on ground-truth benchmark intent datasets.

## Verification Script
Run the deterministic verification checker:
```bash
python3 scripts/check.py examples/synthetic.json
```
Exit codes:
- `0`: Check passed and thresholds satisfied.
- `1`: Threshold breach or operational violation detected.
- `2`: Schema error or invalid usage.

## What the checker recomputes

- `acc` = 100 * correct_selections / total_invocations, then the original `<` check against `min_accuracy_percent`. `correct_selections` cannot exceed `total_invocations`.

The original threshold checks still apply:

- `hallucinated_tools` must not be > `0`.

Every other declared field is type-checked. A negative value is rejected except for deltas, drifts, and gains. A type check does not, by itself, prove the measurement.

`population` and `observedAt` are required. The bundled examples are synthetic fixtures, not market measurements.

```bash
python3 scripts/check.py examples/synthetic.json
python3 scripts/check.py examples/contradiction.json
python3 scripts/check.py examples/broken.json
```

Exit 0 is a pass, exit 1 is a review, and exit 2 is invalid input. A pass does not certify a provider, a regulation, a release, or a price.
