deepgram-dotnet-voice-agent
Use when writing or reviewing C# code in this repo that builds an interactive Deepgram Voice Agent over WebSocket. Covers `ClientFactory.CreateAgentWebSocketClient()`, `SettingsSchema`, event subscriptions, microphone audio streaming, injected user messages, and function-call-related message types. Use `deepgram-dotnet-text-to-speech` for one-way synthesis and STT skills for transcription-only flows.
76
93%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that is highly specific, includes strong trigger terms, clearly answers both what and when, and explicitly distinguishes itself from related skills. The boundary-setting clause ('Use deepgram-dotnet-text-to-speech for one-way synthesis and STT skills for transcription-only flows') is particularly effective for reducing skill selection conflicts. The description uses proper third-person voice throughout.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and components: CreateAgentWebSocketClient(), SettingsSchema, event subscriptions, microphone audio streaming, injected user messages, and function-call-related message types. These are highly specific capabilities. | 3 / 3 |
Completeness | Clearly answers both 'what' (building an interactive Deepgram Voice Agent over WebSocket with specific components) and 'when' ('Use when writing or reviewing C# code in this repo that builds an interactive Deepgram Voice Agent'). Also explicitly delineates boundaries by pointing to other skills for TTS and STT-only flows. | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms: 'C#', 'Deepgram', 'Voice Agent', 'WebSocket', 'microphone audio streaming', 'function-call', plus specific API references like 'ClientFactory.CreateAgentWebSocketClient()' and 'SettingsSchema' that a developer would naturally mention. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: C# + Deepgram + Voice Agent + WebSocket. The description explicitly differentiates from related skills (deepgram-dotnet-text-to-speech for synthesis, STT skills for transcription-only), minimizing conflict risk. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a high-quality skill that efficiently teaches the Deepgram Voice Agent .NET SDK usage with executable code examples, clear structure, and appropriate cross-references. Its main weakness is the lack of explicit validation/error-recovery steps during the streaming phase, though the connection check is a good start. The gotchas section adds significant value by calling out real implementation pitfalls like partial function call support and dynamic provider properties.
Suggestions
Add error handling/validation for the streaming phase (e.g., subscribe to error/close events, show a reconnection pattern or at least mention what happens on disconnect)
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is lean and efficient. It doesn't explain what WebSockets are, what a voice agent is, or how C# works. Every section delivers specific, actionable content. The routing table at the top efficiently directs to other skills. | 3 / 3 |
Actionability | Provides fully executable C# code for authentication, quick start, and microphone streaming. The code is copy-paste ready with real class names, real method signatures, and concrete configuration values. The gotchas section adds critical implementation-specific details. | 3 / 3 |
Workflow Clarity | The quick start shows a clear sequence (create client → subscribe → configure → connect → send → cleanup) with a connection failure check, but there's no explicit validation/error-handling feedback loop for the streaming phase. The microphone streaming section lacks error handling or verification steps. | 2 / 3 |
Progressive Disclosure | Well-structured with clear sections progressing from authentication to quick start to advanced features. References to in-repo files, external docs, and example programs are clearly signaled and one level deep. The routing table at the top efficiently directs users to related skills. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- deepgram/deepgram-dotnet-sdk
- Commit
3b953f0
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.