Name: deepgram
Rating: 59.2 (1 reviews)
Author: membranedev

Blog Docs Log in Get started

deepgram

Deepgram integration. Manage Projects. Use when the user wants to interact with Deepgram data.

Quality

68%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/deepgram/SKILL.md

Quality

Content

78%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.

The content is a strong, executable integration guide with concrete commands, state-handling feedback loops, and clean structure. Its only notable weakness is a padded intro explaining concepts Claude already knows.

Suggestions

Trim the opening paragraph that defines Deepgram and lists business use cases; Claude already knows speech-to-text transcription, so lead directly with how to use the skill.

Formalize the workflow step numbering so the "Skip to Step 2" reference maps to an explicitly numbered Step 2 rather than the mixed 1b/Step labels.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with executable commands, but the opening paragraph explains what Deepgram/speech-to-text is and its business use cases ("call centers, media analysis, and meeting recording") which Claude already knows; this fits mostly-efficient with some unnecessary explanation and is not a 4 due to that padded intro.	3 / 5
Actionability	Fully copy-paste ready bash commands with --json flags, a concrete popular-actions table with real action keys, and an --input JSON example cover the common cases end to end, matching the fully-executable anchor.	5 / 5
Workflow Clarity	A clear sequence (install, login, connect, poll/wait, search, run, proxy) with a state-based feedback loop and re-poll after CLIENT_ACTION_REQUIRED fits clear-sequence-with-most-checkpoints; it is not a 5 because steps are referenced as "Step 2" without formal numbering and labels mix 1b/Step nomenclature.	4 / 5
Progressive Disclosure	A single-file, well-organized guide with clear headers and no nested bundle references; per the simple-skill guidance, well-organized sections with no external references warrant a 5.	5 / 5
	Total	17 / 20 Passed

Description

58%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.

The description answers what and when but relies almost entirely on the brand name and a single action, with no natural trigger terms like transcription or speech-to-text. It is distinct but thin on capability breadth and user-facing keywords.

Suggestions

List more concrete capabilities (e.g., transcribe audio, generate API keys, list models, text-to-speech) so the "what" is comprehensive rather than "Manage Projects" alone.

Add natural trigger terms users would actually say, such as "transcription", "speech-to-text", "audio", or "convert audio to text", instead of only "Deepgram data".

Make the "when" clause more specific than "interact with Deepgram data" by naming the concrete tasks (transcribing, managing projects/keys/models).

Dimension	Reasoning	Score
Specificity	"Manage Projects" is one concrete action while "Deepgram integration" is generic framing, matching the anchor naming a domain with limited concrete actions; it is not a 4 because no further specific actions are enumerated.	3 / 5
Completeness	"Deepgram integration. Manage Projects." states a (thin) what and "Use when the user wants to interact with Deepgram data" supplies a when, fitting the anchor having both with a weaker when; it is not a 5 because the trigger phrases are not concrete and a 3 is ruled out by the explicit when clause.	4 / 5
Trigger Term Quality	The only keywords are the brand "Deepgram" and "interact with Deepgram data", which is technical jargon missing natural phrases a user would say like transcription, speech-to-text, audio, or transcribe; it is above 1 only because the brand term is somewhat relevant.	2 / 5
Distinctiveness Conflict Risk	The brand "Deepgram" plus "Manage Projects" carves a clear niche with distinct triggers and minimal overlap with other skills, matching the clear-niche anchor.	5 / 5
	Total	14 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: membranedev/application-skills
Path: skills/deepgram/SKILL.md
Commit: f484c82

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

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.