CtrlK
BlogDocsLog inGet started
Tessl Logo

deepgram-java-voice-agent

Use when writing or reviewing Java code in this repo that builds an interactive voice agent over `agent.deepgram.com/v1/agent/converse`. Covers `client.agent().v1().v1WebSocket()`, `AgentV1Settings`, `sendSettings`, `sendMedia`, event handlers, provider configuration, and message injection. Use `deepgram-java-text-to-speech` for one-way synthesis or the STT skills for transcription-only flows. Triggers include "voice agent", "agent converse", "full duplex", "barge in", "function call", and "agent websocket".

74

Quality

91%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

82%Weight 40%Scale 1-5

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

A well-structured, highly actionable SDK skill with executable examples and clear cross-skill routing. The main weaknesses are minor verbosity (import block, provider gotcha, promotional footer) and implicit rather than explicit workflow validation checkpoints.

Suggestions

Trim the 11-line Quick-start import block to the few classes actually used in the snippet, or move the full import list to a collapsed note, to improve conciseness.

Convert the inline method/handler bullet lists under 'Key parameters / API surface' into a dedicated references file (or a clearly signaled table) to strengthen progressive disclosure.

Add an explicit verification step after `connect()` (e.g., confirm `onWelcome`/`onSettingsApplied` fired before streaming media) to make the workflow's validation checkpoints explicit.

DimensionReasoningScore

Conciseness

Largely efficient — code blocks and the API-surface list earn their place — but the verbose 11-line import block, the lengthy provider-wrapper gotcha (#5), and the promotional "Central product skills" footer could be trimmed. Not below 3 since most tokens add value; not 5 due to these padded stretches.

4 / 5

Actionability

The Quick start and Message-injection sections are copy-paste-ready executable Java that connect, configure settings on `onWelcome`, wire event handlers, and inject messages, covering the common cases fully.

5 / 5

Workflow Clarity

The connect → register handlers → `onWelcome` → `sendSettings` → stream sequence is shown in code and reinforced by gotcha #2 ("Send settings first"), with a `connect().get(10s)` timeout checkpoint. Not 5 because validation/error-recovery steps are implicit rather than enumerated; the destructive/batch cap does not apply to this connection-setup skill.

4 / 5

Progressive Disclosure

Well-organized into clear sections (When to use, Quick start, Key parameters, API reference, Gotchas) with one-level-deep references to in-repo source paths and external docs, and explicitly notes no `reference.md` exists. Not 5 because the long inline API-surface/handler listing would fit better in a dedicated reference file, and the body exceeds the ~50-line simple-skill exception.

4 / 5

Total

17

/

20

Passed

Description

100%Weight 40%Scale 1-5

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 high-quality description that concretely names the API surface, gives explicit 'Use when' guidance plus a natural trigger list, and actively disambiguates from sibling Deepgram skills. No vague fluff or over-claims.

DimensionReasoningScore

Specificity

Lists multiple concrete actions and a comprehensive API surface — `client.agent().v1().v1WebSocket()`, `AgentV1Settings`, `sendSettings`, `sendMedia`, event handlers, provider configuration, and message injection — rather than vague language.

5 / 5

Completeness

Explicitly answers both what ("writing or reviewing Java code... that builds an interactive voice agent") and when ("Use when...", "Triggers include...") with concrete trigger phrases.

5 / 5

Trigger Term Quality

Explicit trigger list includes natural user phrases — "voice agent", "agent converse", "full duplex", "barge in", "function call", "agent websocket" — covering the common ways a user would request this capability.

5 / 5

Distinctiveness Conflict Risk

Clear niche with explicit de-confliction routing ("Use `deepgram-java-text-to-speech` for one-way synthesis or the STT skills for transcription-only flows"), minimizing overlap with sibling skills.

5 / 5

Total

20

/

20

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.