CtrlK
BlogDocsLog inGet started
Tessl Logo

deepgram-js-audio-intelligence

Use when writing or reviewing JavaScript/TypeScript in this repo that calls Deepgram audio analytics overlays on `/v1/listen` - summarize, topics, intents, sentiment, diarize, redact, detect_language, and entity detection. Same endpoint as plain STT, different params. Covers REST via `client.listen.v1.media.transcribeUrl` / `transcribeFile` and the WebSocket-supported subset on `client.listen.v1.createConnection()` / `connect()`. Use `deepgram-js-speech-to-text` for plain transcription and `deepgram-js-text-intelligence` for analytics on already-transcribed text. Triggers include "audio intelligence", "summarize audio", "diarize", "sentiment from audio", "redact PII", and "detect language audio".

75

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

85%

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

A well-structured, actionable reference with executable code and clear navigation. Its main weakness is conciseness: availability facts are repeated across the table, parameter list, and gotchas, and the inline docs-URL list is heavy.

Suggestions

Collapse the redundant REST-vs-WSS availability info: keep either the feature table or gotcha #2 and remove the restatement, so the REST-only flag list (topics, intents, sentiment, summarize, detect_language) is not repeated three times.

Move the nine product-doc URLs into a single referenced reference file or trim to the two most essential, keeping the in-repo reference.md and OpenAPI/AsyncAPI links as the primary pointers.

Reconcile gotcha #1 (summarize is versioned and points at 'v2') with the Quick start code that passes `summarize: true`, so the headline example matches the versioned form it describes.

DimensionReasoningScore

Conciseness

The body is dense and free of concept-explanation padding, but the REST-vs-WSS availability facts are restated across the feature table, the key-parameters list, and gotcha #2, and the nine inline product-doc URLs are a heavy token cost that could be tightened.

2 / 3

Actionability

Provides complete, executable REST (transcribeUrl with all overlay params) and WSS (createConnection) code blocks with concrete method names and named example files, making it copy-paste ready rather than pseudocode or missing key details.

3 / 3

Workflow Clarity

The primary REST path is a single unambiguous call and the body is clearly sequenced (when-to-use, availability, auth, quick starts, params, gotchas); as a non-destructive API-call skill it needs no validation checkpoints, so the simple-skill allowance applies.

3 / 3

Progressive Disclosure

Well-organized into clearly signaled one-level-deep references (reference.md, in-repo example files, OpenAPI/AsyncAPI/Context7, product docs) with no nested pointer chains; no bundle files exist to verify, but the in-repo paths and links are cleanly navigable.

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 strong, concrete description that names specific capabilities, API surfaces, and natural trigger phrases while explicitly distinguishing sibling skills. It fully answers both what and when with no vague fluff.

DimensionReasoningScore

Specificity

Names concrete actions (writing/reviewing JS/TS, calling named analytics overlays) and specific method surfaces (transcribeUrl/transcribeFile, createConnection/connect), matching the 'multiple specific concrete actions' anchor; it is not the level below because it goes well beyond naming only a domain and some actions.

3 / 3

Completeness

Clearly states what it does (analytics overlays on /v1/listen with named features) and when to use it ('Use when writing or reviewing...'), plus an explicit 'Triggers include...' clause, so both what and when are explicit rather than implied.

3 / 3

Trigger Term Quality

Explicit natural triggers ('summarize audio', 'diarize', 'sentiment from audio', 'redact PII', 'detect language audio') are phrases users would actually say, with good coverage of common variations rather than just a few relevant keywords.

3 / 3

Distinctiveness Conflict Risk

Explicitly routes sibling cases to deepgram-js-speech-to-text (plain transcription) and deepgram-js-text-intelligence (analytics on already-transcribed text), giving it a clear niche unlikely to trigger for the wrong skill.

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.