Loading ToolBench
A BLE (Bluetooth Low Energy) edge agent that monitors BLE devices, persists telemetry, and acts autonomously. Includes device presence monitoring, plugin system, and dual-agent architecture (controller and user-facing).
All three tools have empty input schemas (no properties, no required fields) and minimal descriptions (10-30 chars). No parameter descriptions exist. Tool names follow verb_noun pattern but lack specificity. No output schemas documented. No error handling guidance. This server falls well below production baseline (median ~50). The device_monitor plugin shows domain logic but tool definitions are skeletal.
Start the continuous BLE device presence monitor.
Return current presence state for all monitored devices.
Stop the device presence monitor.
All tools have empty input schemas: inputSchema={"type":"object","properties":{},"required":[]}.
Tool descriptions are 10-30 characters, below the 34-char p10 baseline. 'Start the continuous BLE device presence monitor' (51 chars) is acceptable, but 'Stop the device presence monitor' (31 chars) and 'Return current presence state for all monitored devices' (54 chars) lack context on WHEN to use, WHAT it returns, and prerequisites.
No output schemas documented. LLMs cannot plan downstream calls or extract data without knowing what device_monitor.status returns (e.g., does it return {devices: [{name, present, last_seen_ts}]} or a flat list?).
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 31 | <=2025-11-25 | v2 |
No error handling guidance. If device_monitor.start fails (e.g., BLE adapter unavailable), the tool provides no recovery hint. Per pattern:recovery-guide, errors must tell the LLM what to do next.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). device_monitor.start and device_monitor.stop are WRITE operations but lack destructiveHint=true. device_monitor.status is READ_ONLY but lacks readOnlyHint=true. Per current spec (2026-07-28), these annotations guide agent planning.