deepgram-dotnet-text-to-speech
Use when writing or reviewing C# code in this repo that calls Deepgram Text-to-Speech. Covers `ClientFactory.CreateSpeakRESTClient()` with `ToStream` / `ToFile`, and `ClientFactory.CreateSpeakWebSocketClient()` with `Connect`, `SpeakWithText`, `Flush`, and streaming `AudioResponse` events. Use `deepgram-dotnet-voice-agent` for full-duplex assistants instead of one-way synthesis.
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 precisely identifies the technology stack (C#, Deepgram TTS), lists specific API methods and patterns, and provides clear 'Use when' guidance. The disambiguation clause separating one-way synthesis from full-duplex assistants is a strong addition that minimizes conflict with related skills. The description is concise yet comprehensive.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and API methods: CreateSpeakRESTClient with ToStream/ToFile, CreateSpeakWebSocketClient with Connect/SpeakWithText/Flush, and streaming AudioResponse events. Also distinguishes from a related skill for full-duplex assistants. | 3 / 3 |
Completeness | Clearly answers both 'what' (covers REST and WebSocket TTS clients with specific methods) and 'when' (opens with 'Use when writing or reviewing C# code in this repo that calls Deepgram Text-to-Speech'). Also includes a disambiguation clause directing users to a different skill for full-duplex assistants. | 3 / 3 |
Trigger Term Quality | Includes highly relevant natural keywords: 'C#', 'Deepgram', 'Text-to-Speech', 'SpeakRESTClient', 'SpeakWebSocketClient', 'ToStream', 'ToFile', 'streaming', 'AudioResponse', 'one-way synthesis'. These are terms a developer would naturally use when working with this SDK. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: C# + Deepgram + Text-to-Speech + specific API methods. The explicit disambiguation from 'deepgram-dotnet-voice-agent' for full-duplex assistants further reduces conflict risk with related skills. | 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 covers both REST and WebSocket TTS patterns with executable code, clear organization, and useful gotchas that address real pitfalls (method naming, flush behavior, format matching). The only notable gap is the lack of explicit error recovery or validation steps in the WebSocket streaming workflow, where audio data could silently fail to arrive. Overall, it's concise, actionable, and well-structured.
Suggestions
Add a brief validation/verification step to the WebSocket workflow (e.g., check that bytes were actually written to the output file, or log byte count after the session ends) to strengthen error recovery for streaming operations.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is lean and efficient. It doesn't explain what TTS is or how WebSockets work—it assumes Claude's competence. Every section delivers specific, actionable information. The comments in code are minimal and purposeful (e.g., explaining the raw PCM format gotcha). | 3 / 3 |
Actionability | Both REST and WebSocket quick starts provide fully executable, copy-paste-ready C# code with correct using statements, initialization, and complete workflows. The Gotchas section provides specific method names and concrete behavioral details rather than vague warnings. | 3 / 3 |
Workflow Clarity | The REST flow is straightforward and clear. The WebSocket flow has a good sequence (Subscribe → Connect → SpeakWithText → Flush → Stop) with a connection check, but lacks explicit validation/error-handling for the audio output path (e.g., verifying bytes were received, handling partial failures). For a streaming operation where data could be lost, a feedback loop or verification step would strengthen this. | 2 / 3 |
Progressive Disclosure | The skill is well-structured with clear sections (When to use, Authentication, Quick starts, Key params, References, Gotchas, Example files). It references in-repo source files, example programs, external docs, and a related skill (`deepgram-dotnet-voice-agent`) all at one level deep with clear signaling. Content is appropriately sized for inline presentation. | 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.