CtrlK
BlogDocsLog inGet started
Tessl Logo

deepgram-go-audio-intelligence

Use when writing or reviewing Go code in this repo that applies summaries, topics, intents, sentiment, language detection, diarization, redaction, or entity extraction to audio inputs through Listen v1 REST. Route plain transcription to deepgram-go-speech-to-text and plain-text Read requests to deepgram-go-text-intelligence.

75

Quality

92%

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.

This is a well-structured SDK skill with an excellent executable quick start example and clear parameter documentation. The progressive disclosure is strong with layered references to in-repo code, external APIs, and documentation. Minor verbosity in the API reference links and the promotional 'Central product skills' section slightly reduce token efficiency.

Suggestions

Trim the API reference section to only the most essential references (in-repo types file and the main product docs overview page), moving the exhaustive list to a separate reference file if needed.

Remove or condense the 'Central product skills' promotional block, which adds tokens without directly aiding task completion.

DimensionReasoningScore

Conciseness

Generally efficient but includes some unnecessary sections like the 'Central product skills' promo block and the 'Gotchas' section partially repeats information already conveyed (REST-only nature). The API reference section is quite long with many links that could be trimmed.

2 / 3

Actionability

The quick start provides a fully executable, copy-paste-ready Go program with correct imports, client initialization, option configuration, and response handling. Key parameters are concretely named with their types/values.

3 / 3

Workflow Clarity

This is essentially a single-task skill (make a REST call with analytics flags). The workflow is unambiguous: set env var → create client → call FromURL with options → read response. For this level of complexity, the clarity is sufficient and no destructive/batch operations require validation checkpoints.

3 / 3

Progressive Disclosure

Content is well-structured with a concise quick start up front, key parameters listed separately, and a layered API reference section pointing to in-repo files, OpenAPI specs, and product docs — all one level deep with clear signaling. Example files are referenced by path.

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.

This is an excellent skill description that clearly defines its scope (audio intelligence features via Listen v1 REST in Go), provides explicit trigger conditions, and actively disambiguates itself from related skills by specifying routing rules for adjacent use cases. The description is concise yet comprehensive, with strong natural trigger terms and clear boundaries.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions: summaries, topics, intents, sentiment, language detection, diarization, redaction, entity extraction applied to audio inputs through Listen v1 REST. Also specifies routing rules for plain transcription and plain-text Read requests.

3 / 3

Completeness

Clearly answers both what (applies audio intelligence features via Listen v1 REST) and when ('Use when writing or reviewing Go code in this repo that applies summaries, topics, intents...'). Also explicitly states when NOT to use it by routing other cases to different skills.

3 / 3

Trigger Term Quality

Includes strong natural keywords users would say: 'Go code', 'summaries', 'topics', 'intents', 'sentiment', 'language detection', 'diarization', 'redaction', 'entity extraction', 'audio', 'Listen v1 REST', plus explicit routing terms like 'deepgram-go-speech-to-text' and 'deepgram-go-text-intelligence'.

3 / 3

Distinctiveness Conflict Risk

Highly distinctive with explicit boundary-drawing: routes plain transcription to deepgram-go-speech-to-text and plain-text Read requests to deepgram-go-text-intelligence, clearly carving out its niche as audio intelligence features only via Listen v1 REST in Go.

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-go-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.