CtrlK
BlogDocsLog inGet started
Tessl Logo

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".

72

Quality

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Advisory

Suggest reviewing before use

SKILL.md
Quality
Evals
Security

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-organized Java SDK skill with executable code and a strong gotchas list. It is slightly verbose in places and lacks local reference files and explicit verification checkpoints.

Suggestions

Tighten the REST visitor example by extracting the null-guarding channel/alternative traversal into a brief helper note rather than inlining the full verbose block.

Add explicit validation checkpoints (e.g., verify the transcript is non-empty / check WebSocket connect().get() succeeded before sending media) to lift workflow clarity.

Move the long REST/WSS parameter inventories into a local references file (e.g., reference.md) and link to it from the body to improve progressive disclosure.

DimensionReasoningScore

Conciseness

Mostly efficient and free of generic concept explanations, but the inline REST response visitor block (null-guarding channels/alternatives) and the long API-surface param list add tokens that lean beyond what Claude strictly needs.

2 / 3

Actionability

Provides complete, copy-paste-ready Java for REST URL, file bytes, WebSocket streaming, and async, plus Gradle/Maven deps and explicit method/handler inventories.

3 / 3

Workflow Clarity

Sections are well organized and gotchas sequence WebSocket setup (register handlers before connect, wait on connect().get()), but there are no explicit validate/verify checkpoints for the transcription outputs or batch operations.

2 / 3

Progressive Disclosure

Good section structure, but no bundle files exist; the 'API reference (layered)' and 'Example files' sections point to external/in-repo sources rather than one-level-deep local reference files, so content that could be split is inline.

2 / 3

Total

9

/

12

Passed

Description

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.

A precise, third-person description that names concrete SDK surfaces, gives natural trigger terms, and disambiguates against sibling skills. It answers both what and when with no padding.

DimensionReasoningScore

Specificity

Lists multiple concrete actions and surface names: 'transcribeUrl / transcribeFile (REST)' and 'v1WebSocket() (WebSocket)' for prerecorded and live transcription.

3 / 3

Completeness

Explicitly answers both what (REST/WebSocket transcription on /v1/listen) and when, opening with 'Use when writing or reviewing Java code...' plus a Triggers list.

3 / 3

Trigger Term Quality

Good coverage of natural terms users would say: 'transcribe', 'speech to text', 'STT', 'listen v1', 'nova-3', 'live transcription', and 'websocket transcription'.

3 / 3

Distinctiveness Conflict Risk

Clear niche (Java SDK on /v1/listen) with explicit redirects to sibling skills (audio-intelligence, conversational-stt, voice-agent), making wrong-skill triggering unlikely.

3 / 3

Total

12

/

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
deepgram/deepgram-java-sdk
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.