deepgram-js-speech-to-text
Use when writing or reviewing JavaScript/TypeScript in this repo that calls Deepgram Speech-to-Text v1 (`/v1/listen`) for prerecorded or live audio transcription. Covers `client.listen.v1.media.transcribeUrl` / `transcribeFile` (REST) plus `client.listen.v1.createConnection()` / `connect()` (WebSocket). Use `deepgram-js-audio-intelligence` for summarize/sentiment/topics/diarize overlays, `deepgram-js-conversational-stt` for Flux turn-taking on `/v2/listen`, and `deepgram-js-voice-agent` for full-duplex assistants. Triggers include "transcribe", "speech to text", "STT", "listen.v1", "nova-3", "live transcription", and "websocket transcription".
94
92%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
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 hits all rubric dimensions at the highest level. It provides specific API methods and endpoints, includes a comprehensive set of natural trigger terms, clearly states both what the skill does and when to use it, and proactively distinguishes itself from related skills by naming them and their scopes. The explicit boundary-drawing with sibling skills is a particularly strong feature.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and API methods: transcribeUrl, transcribeFile (REST), createConnection/connect (WebSocket), and clearly names the exact endpoints (/v1/listen) and SDK patterns. | 3 / 3 |
Completeness | Clearly answers both 'what' (Deepgram STT v1 for prerecorded/live audio transcription via REST and WebSocket) and 'when' (explicit 'Use when' clause at the start, plus explicit trigger terms and boundary conditions distinguishing from related skills). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms explicitly listed: 'transcribe', 'speech to text', 'STT', 'listen.v1', 'nova-3', 'live transcription', 'websocket transcription'. These are terms users would naturally use when needing this skill. | 3 / 3 |
Distinctiveness Conflict Risk | Exceptionally distinctive — explicitly delineates boundaries by naming sibling skills (deepgram-js-audio-intelligence, deepgram-js-conversational-stt, deepgram-js-voice-agent) and their respective domains, making it very clear when NOT to use this skill. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, well-structured skill that provides executable code examples for all primary use cases, clear workflow sequencing for the WebSocket path, and excellent progressive disclosure with layered references. Minor verbosity in the introductory routing section and some explanatory phrases could be trimmed, but overall the content is highly actionable and well-organized.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient but includes some unnecessary explanation (e.g., 'Good for batch jobs, caption generation, offline processing' and 'Good for live captions, microphone audio, telephony streams' are somewhat redundant for Claude). The 'When to use this product' section and cross-references to other skills add useful routing but could be tighter. The key parameters list is a useful dense reference. | 2 / 3 |
Actionability | Provides fully executable, copy-paste-ready code examples for all three main use cases (REST URL, REST file, WebSocket). Includes specific method names, event types, and client messages. The gotchas section provides concrete, specific guidance rather than vague warnings. | 3 / 3 |
Workflow Clarity | The WebSocket workflow clearly sequences the two-stage flow: createConnection → register handlers → connect → waitForOpen → sendMedia → sendFinalize. The gotchas section explicitly calls out validation-like concerns (finalize before closing, keep-alive for idle streams, encoding must match). For the REST path, the single-step nature is unambiguous. | 3 / 3 |
Progressive Disclosure | Well-structured with quick starts up front, then layered API references pointing to in-repo files, OpenAPI/AsyncAPI specs, and external docs. References are one level deep and clearly signaled. The example files listing and cross-skill routing ('Use a different skill when') provide excellent navigation. | 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-js-sdk
- Commit
c567b98
- 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.