deepgram-java-speech-to-text
Use when writing or reviewing Java code in this repo that calls Deepgram Speech-to-Text v1 (`/v1/listen`) for prerecorded or live transcription. Covers `client.listen().v1().media().transcribeUrl` / `transcribeFile` (REST) and `client.listen().v1().v1WebSocket()` (WebSocket). Use `deepgram-java-audio-intelligence` for analytics overlays, `deepgram-java-conversational-stt` for Flux `/v2/listen`, and `deepgram-java-voice-agent` for full-duplex assistants. Triggers include "transcribe", "speech to text", "STT", "listen v1", "nova-3", "live transcription", and "websocket transcription".
89
86%
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 clearly defines its scope (Deepgram STT v1 in Java), lists concrete API methods and actions, provides explicit trigger terms, and proactively distinguishes itself from related skills. The 'Use when' clause is present and well-formed, and the boundary-setting with sibling skills is a strong differentiator that reduces conflict risk.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: writing/reviewing Java code, calling Deepgram STT v1 `/v1/listen`, prerecorded and live transcription, specific API methods (`transcribeUrl`, `transcribeFile`, `v1WebSocket`), and distinguishes REST vs WebSocket approaches. | 3 / 3 |
Completeness | Clearly answers both 'what' (Java code for Deepgram STT v1 prerecorded/live transcription via REST and WebSocket) and 'when' (explicit 'Use when' clause at the start plus explicit trigger terms at the end). Also clarifies when NOT to use it by pointing to sibling skills. | 3 / 3 |
Trigger Term Quality | Explicitly lists natural trigger terms users would say: 'transcribe', 'speech to text', 'STT', 'listen v1', 'nova-3', 'live transcription', 'websocket transcription'. These cover common variations and natural language a developer would use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive: scoped to a specific API version (`/v1/listen`), specific language (Java), and specific vendor (Deepgram). Explicitly delineates boundaries by naming sibling skills for analytics, v2/Flux, and voice agent use cases, minimizing overlap risk. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
72%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-organized skill that provides fully actionable Java code examples for Deepgram STT v1 across REST and WebSocket patterns. Its main weakness is the lack of error handling/validation guidance and some verbosity in the code examples (particularly the visitor pattern boilerplate). The progressive disclosure and cross-referencing to related skills and documentation is excellent.
Suggestions
Add a brief error handling example or validation checkpoint (e.g., try/catch around connect/transcribe with common failure modes like invalid API key or encoding mismatch) to improve workflow clarity.
Consider trimming the REST URL visitor example—extract the transcript-access pattern into a helper method or show the concise happy-path first, with the full null-guarded version as a note.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient with good code examples, but includes some unnecessary explanations (e.g., explaining what Token vs Bearer auth means, the note about transcribeFile reading into memory) and the code examples are quite verbose—particularly the REST URL example with the full visitor pattern and null-checking boilerplate. The 'Key parameters / API surface' section is a useful reference but borders on exhaustive listing. | 2 / 3 |
Actionability | All code examples are fully executable with correct imports, concrete builder patterns, and real API calls. The REST URL, REST file, WebSocket, and async examples are all copy-paste ready with specific class names, methods, and patterns. The dependency declarations (Gradle/Maven) are also concrete and versioned. | 3 / 3 |
Workflow Clarity | The skill presents clear sequences for individual operations (REST call, WebSocket connect/send/close) and the WebSocket section correctly sequences handler registration before connect. However, there are no explicit validation or error-handling checkpoints—no try/catch patterns, no guidance on what to do when transcription fails, and no feedback loops for common failure modes like auth errors or encoding mismatches. | 2 / 3 |
Progressive Disclosure | The skill is well-structured with clear sections progressing from authentication to quick starts to advanced topics. It appropriately references other skills (audio-intelligence, conversational-stt, voice-agent), lists example files in the repo, and provides layered API references (in-repo source, OpenAPI, AsyncAPI, docs). The 'When to use this product' section effectively routes users to the right skill. | 3 / 3 |
Total | 10 / 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-java-sdk
- Commit
de2dd4b
- 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.