CtrlK
BlogDocsLog inGet started
Tessl Logo

deepgram-java-conversational-stt

Use when writing or reviewing Java code in this repo that calls Deepgram Conversational STT v2 / Flux over `/v2/listen`. Covers `client.listen().v2().v2WebSocket()`, `V2ConnectOptions`, `onTurnInfo`, and turn-aware close handling. Use `deepgram-java-speech-to-text` for standard v1 transcription and `deepgram-java-voice-agent` for fully interactive assistants. Triggers include "flux", "conversational stt", "listen v2", "turn detection", "end of turn", and "eot".

76

Quality

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Quality

Content

87%

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

A tight, executable Java SDK skill with complete quick-start code, a high-signal gotchas section, and cleanly layered references. Its main weakness is that the streaming connect workflow is conveyed implicitly across code and gotchas rather than as an explicitly sequenced, validated flow.

Suggestions

Add a short numbered 'Stream lifecycle' workflow (1. connect+get, 2. wire handlers before sending, 3. sendMedia, 4. sendCloseStream) so the sequencing and the 'wait for connect before media' checkpoint live in one place.

Add an explicit validation/checkpoint step for the connect handshake (e.g. confirm onConnected/request_id before streaming) to give the workflow a clear feedback point.

Collapse the small 'When to use this product' overlap with the frontmatter description to avoid restating scope twice.

DimensionReasoningScore

Conciseness

Lean body that assumes Claude's competence — 'Turn-aware streaming transcription over `/v2/listen`' with no padding explaining what STT or WebSockets are; gotchas are terse and high-signal.

3 / 3

Actionability

Quick start gives fully executable Java code (connect, onTurnInfo handler, sendCloseStream) plus a verified list of exact connect options, all copy-paste ready rather than pseudocode.

3 / 3

Workflow Clarity

The connect→handlers→media→close sequence is present and one ordering checkpoint is explicit ('Wait for `connect(...).get(...)` before sending media'), but the flow is implicit across code and a gotcha rather than a numbered sequence with explicit validation checkpoints.

2 / 3

Progressive Disclosure

Well-organized sections with clearly signaled one-level-deep references (in-repo source path, AsyncAPI, Context7, product docs), and it explicitly notes no nested reference.md exists in this checkout.

3 / 3

Total

11

/

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, well-scoped description that names concrete API surfaces, provides natural trigger terms, and explicitly distinguishes itself from adjacent skills. It uses third-person voice throughout with no vague fluff.

DimensionReasoningScore

Specificity

Lists multiple concrete API surfaces — `client.listen().v2().v2WebSocket()`, `V2ConnectOptions`, `onTurnInfo`, and 'turn-aware close handling' — rather than vague actions.

3 / 3

Completeness

Explicitly answers both what (calling Deepgram Conversational STT v2 / Flux over `/v2/listen`) and when via an explicit 'Use when...' clause plus named triggers.

3 / 3

Trigger Term Quality

Curated natural trigger list ('flux', 'conversational stt', 'listen v2', 'turn detection', 'end of turn', 'eot') covers the common phrasings a user would say.

3 / 3

Distinctiveness Conflict Risk

Explicitly disambiguates from sibling skills (`deepgram-java-speech-to-text` for v1, `deepgram-java-voice-agent` for assistants), giving it a clear niche unlikely to conflict.

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.