CtrlK
BlogDocsLog inGet started
Tessl Logo

deepgram-js-voice-agent

Use when writing or reviewing JavaScript/TypeScript in this repo that builds an interactive voice agent via `agent.deepgram.com/v1/agent/converse`. Covers `client.agent.v1.createConnection()` / `connect()`, `sendSettings`, `sendMedia`, runtime updates, event handling, and function-call responses. Use `deepgram-js-text-to-speech` for one-way synthesis, `deepgram-js-speech-to-text` or `deepgram-js-conversational-stt` for transcription only, and `deepgram-js-management-api` for project/model admin rather than live agent runtime. Triggers include "voice agent", "agent converse", "full duplex", "barge-in", "function calling", and "agent.v1".

75

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

85%

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

The body is a dense, actionable reference with executable quick-start code, a clear connect workflow, and well-organized external/in-repo pointers. Its main weakness is mild token redundancy where the body repeats the frontmatter's when-to-use and sibling-skill guidance.

Suggestions

Trim or remove the 'When to use this product' and 'Use a different skill when' sections — that guidance already lives in the frontmatter description, so restating it in the body wastes tokens.

Cut the promotional 'Central product skills' section (`npx skills add deepgram/skills`) unless it is essential to the task; it reads as marketing rather than actionable guidance.

DimensionReasoningScore

Conciseness

The body is mostly lean and assumes competence, but the "When to use this product" / "Use a different skill when" block and the "Central product skills" promo largely restate the frontmatter's when and sibling-skill guidance, so it could be tightened.

2 / 3

Actionability

The Quick start gives fully executable code — `createConnection()`, `connect()`, `waitForOpen()`, and a complete `sendSettings({...})` payload — with concrete method names and copy-paste-ready detail, matching the score-3 anchor.

3 / 3

Workflow Clarity

The connect flow is clearly sequenced (create connection → on message → connect → waitForOpen → sendSettings) with an explicit readiness checkpoint and ordering gotchas such as "Settings must be first," giving an unambiguous sequence.

3 / 3

Progressive Disclosure

Content is organized into well-labeled sections with clearly signaled one-level references (in-repo `reference.md`, `src/CustomClient.ts`, `examples/*.ts`, and external OpenAPI/AsyncAPI URLs); there are no nested reference chains, and no bundle files are needed.

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.

The description is a strong, concrete trigger: it names the live voice-agent runtime, lists specific SDK methods, gives explicit 'Use when' guidance, and disambiguates from four sibling skills. It closely matches the rubric's good examples on every dimension.

DimensionReasoningScore

Specificity

Quotes "Covers `client.agent.v1.createConnection()` / `connect()`, `sendSettings`, `sendMedia`, runtime updates, event handling, and function-call responses" — it lists multiple specific concrete SDK actions rather than vague language, matching the score-3 anchor.

3 / 3

Completeness

It opens with "Use when writing or reviewing JavaScript/TypeScript in this repo that builds an interactive voice agent" (when) and follows with the concrete API coverage (what), explicitly answering both what and when with explicit triggers.

3 / 3

Trigger Term Quality

Explicit trigger list "voice agent", "agent converse", "full duplex", "barge-in", "function calling", and "agent.v1" gives good coverage of terms a user would naturally say; it is not merely internal jargon.

3 / 3

Distinctiveness Conflict Risk

It carves a clear live-agent-runtime niche and redirects to siblings ("Use `deepgram-js-text-to-speech` for one-way synthesis ... `deepgram-js-management-api` for project/model admin rather than live agent runtime"), making conflict with adjacent skills 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-js-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.