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".

71

Quality

86%

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

72%

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

This is a solid, well-organized skill that provides concrete, executable guidance for using Deepgram's audio intelligence features. Its main strengths are the clear REST vs WSS feature availability table, actionable code examples, and excellent cross-skill routing. Weaknesses include missing error handling/validation workflows and some token inefficiency in the reference links and authentication sections.

Suggestions

Add a brief error handling or response validation step after the REST quick start (e.g., checking that `data.results.summary` exists when `summarize` is enabled) to improve workflow clarity.

Trim the authentication section — Claude knows how to initialize an SDK client with an API key from env vars — and consolidate the 8 product doc links into a single reference line.

DimensionReasoningScore

Conciseness

Generally efficient but includes some unnecessary content like the 'Central product skills' section and the lengthy list of product doc links. The feature availability table and gotchas are valuable, but the API reference section with 8 documentation links could be more compact. The authentication section explains something Claude likely knows how to do.

2 / 3

Actionability

Provides fully executable, copy-paste ready code for both REST and WSS paths. The parameter lists are concrete and specific, the quick start examples show real API calls with actual parameter values, and the gotchas section provides specific, actionable warnings (e.g., use `keyterm` not `keywords` for Nova-3).

3 / 3

Workflow Clarity

The skill clearly separates REST vs WSS paths and provides a feature availability table, but there are no validation or verification steps. For operations involving audio processing with multiple analytics overlays, there's no guidance on checking results, handling errors, or verifying that intelligence features returned expected data structures.

2 / 3

Progressive Disclosure

Well-structured with clear sections progressing from when-to-use routing, to quick starts, to parameters, to layered API references, to gotchas. References to other skills, example files, and external docs are clearly signaled and one level deep. The routing table at the top effectively directs users to the right skill.

3 / 3

Total

10

/

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.

This is an excellent skill description that is highly specific, includes strong trigger terms, clearly delineates both what the skill does and when to use it, and proactively distinguishes itself from related skills. The boundary guidance with sibling skills is particularly well done, reducing conflict risk. The only minor concern is the density of information which could be slightly more structured, but the content itself is comprehensive and precise.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions and features: summarize, topics, intents, sentiment, diarize, redact, detect_language, entity detection. Also specifies exact API methods like `client.listen.v1.media.transcribeUrl`, `transcribeFile`, and `createConnection()`/`connect()`.

3 / 3

Completeness

Clearly answers both 'what' (audio analytics overlays on /v1/listen including summarize, topics, intents, sentiment, etc.) and 'when' (explicit 'Use when writing or reviewing JavaScript/TypeScript... that calls Deepgram audio analytics overlays'). Also provides boundary guidance distinguishing from related skills (`deepgram-js-speech-to-text` and `deepgram-js-text-intelligence`).

3 / 3

Trigger Term Quality

Includes a rich set of natural trigger terms users would say: 'audio intelligence', 'summarize audio', 'diarize', 'sentiment from audio', 'redact PII', 'detect language audio'. Also includes technical terms like specific method names and feature names that developers would reference.

3 / 3

Distinctiveness Conflict Risk

Explicitly differentiates itself from two related skills by name ('deepgram-js-speech-to-text' for plain transcription and 'deepgram-js-text-intelligence' for text analytics), carving out a clear niche for audio analytics overlays specifically. The distinction between 'same endpoint, different params' is very helpful for disambiguation.

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.

Validation11 / 11 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.