genpark-voice-turn-taking-endpoint-detector-skill
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| analyze_turn_statusD | analyze turn status |
| calibrate_acoustic_thresholdsD | calibrate acoustic thresholds |
| predict_semantic_closureD | predict semantic closure |
| run_benchmark_turn_detectionC | run benchmark turn detection |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
Each tool targets a clearly distinct action: analyze turn status, calibrate thresholds, predict semantic closure, and run benchmarks. There is no overlapping purpose or ambiguous boundary between them.
All names follow a consistent snake_case verb_noun pattern (analyze_, calibrate_, predict_, run_). The convention is predictable and easy to parse.
Four tools is well-scoped for a focused endpoint-detection skill, covering analysis, calibration, prediction, and benchmarking without bloat.
The surface covers core operations: status analysis, threshold calibration, semantic closure prediction, and benchmarking. A minor gap exists for real-time streaming detection or explicit audio input handling, but agents can work around it.