CtrlK
BlogDocsLog inGet started
Tessl Logo

deepgram-js-conversational-stt

Use when writing or reviewing JavaScript/TypeScript in this repo that calls Deepgram Conversational STT v2 / Flux (`/v2/listen`) for turn-aware streaming transcription. Covers `client.listen.v2.createConnection()` / `connect()`, Flux models, and turn events like `TurnInfo`. Use `deepgram-js-speech-to-text` for standard v1 ASR and `deepgram-js-voice-agent` for full-duplex assistants. Triggers include "flux", "v2 listen", "conversational STT", "turn detection", "end of turn", "EOT", and "listen.v2".

76

Quality

95%

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

90%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 tight, repo-anchored skill body: executable code, precise source-file-cited API surface, and high-signal gotchas with no padding. Workflow sequencing is clear with an implicit error path, and references are layered and well-signaled though the absence of any bundle files means more detail stays inline than ideal.

Suggestions

Add an explicit error-recovery checkpoint in the quick start (e.g., on FatalError, log the payload and reconnect with backoff) to turn the implicit error path into a clear validate/retry loop.

Consider extracting the connect-argument and socket-method tables into a references/ file so the body stays a pure overview and progressive disclosure can reach a fully one-level-deep structure.

Surface one short inline note clarifying which connect args are required vs optional, so the executable example is self-sufficient without opening the cited source files.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence: it skips generic explanations of what STT or websockets are and delivers only repo-specific surface, params, gotchas, and a single working example. Every section earns its place.

5 / 5

Actionability

Provides a copy-paste-ready quick start with real method names, a concrete connect/handler/sendMedia/sendCloseStream flow, and a precise API surface map with source file paths; the gotchas give executable do/don't rules (use sendCloseStream not sendFinalize, omit encoding for containerized audio).

5 / 5

Workflow Clarity

The quick start sequences createConnection → register handlers → connect → waitForOpen → sendMedia → sendCloseStream with a FatalError close path, which is a clear sequence with an implicit error checkpoint. It lacks an explicit validate/retry feedback loop, but this is a non-destructive read-only streaming skill, so the destructive/batch cap does not apply.

4 / 5

Progressive Disclosure

Well-organized into labeled sections (Quick start, Key parameters, Limitations, API reference layered, Gotchas, Example files) with a clearly signaled layered reference list pointing to canonical OpenAPI/AsyncAPI/docs. No bundle files exist to offload detail into, so some reference material (e.g. the full connect-arg list) is inline rather than split into a separate file.

4 / 5

Total

18

/

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.

An exemplary description: it states a precise capability scope, lists natural trigger terms, gives explicit when-to-use guidance, and actively disambiguates from related skills. It uses third person throughout and avoids fluff.

DimensionReasoningScore

Specificity

Names concrete, specific actions and the exact API surface (`createConnection()` / `connect()`, Flux models, TurnInfo turn events), plus concrete trigger terms; coverage of the skill's capabilities is comprehensive.

5 / 5

Completeness

Opens with an explicit 'Use when...' trigger clause and pairs it with a clear statement of what the skill does, plus explicit redirects to adjacent skills; both what and when are concrete and explicit.

5 / 5

Trigger Term Quality

Includes natural user-facing phrases and synonyms (`flux`, `v2 listen`, `conversational STT`, `turn detection`, `end of turn`, `EOT`, `listen.v2`) that a user would plausibly say, with broad keyword coverage.

5 / 5

Distinctiveness Conflict Risk

The description carves out a narrow niche (turn-aware v2/Flux STT) and explicitly disambiguates from sibling skills (`deepgram-js-speech-to-text`, `deepgram-js-voice-agent`, `deepgram-js-audio-intelligence`), minimizing mis-trigger risk.

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-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.