deepgram-performance-tuning

Optimize Deepgram API performance for faster transcription and lower latency. Use when improving transcription speed, reducing latency, or optimizing audio processing pipelines. Trigger: "deepgram performance", "speed up deepgram", "optimize transcription", "deepgram latency", "deepgram faster", "deepgram throughput".

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./plugins/saas-packs/deepgram-pack/skills/deepgram-performance-tuning/SKILL.md

Quality

Content

65%Weight 40%Scale 1-3

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

The content is highly actionable with executable code and concrete parameters, but it is verbose for the context window, omits validation feedback loops for batch/destructive operations, and fails to use the available reference bundle, leaving full implementations inline.

Suggestions

Move the six full TypeScript implementations into references/implementation.md and link to them from SKILL.md, keeping only the concise Performance Levers table and a short quick-start example inline.

Add explicit validation checkpoints to the batch and preprocessing workflows — e.g., verify the preprocessed WAV with ffprobe before transcription, and add a validate→retry loop when 429s occur during batch processing.

Trim debug console.log statements and pricing comments from the inline code to reduce token usage while preserving the actionable core.

Dimension	Reasoning	Score
Conciseness	The prose (Overview, Performance Levers table, Error Handling) is lean and assumes Claude's competence, but the body carries six full TypeScript implementations with debug console.log lines and pricing annotations ("Nova-3: Best accuracy, fast, $0.0043/min") that could be tightened, so it is mostly efficient rather than lean throughout.	2 / 3
Actionability	It provides fully executable ffmpeg commands with specific flags, complete TypeScript functions, concrete model names, fixed concurrency (50), and Redis caching with a TTL — copy-paste ready guidance matching the anchor-3 example.	3 / 3
Workflow Clarity	Steps 1–6 are clearly sequenced and the Error Handling table gives reactive remedies, but batch processing (Step 4) and audio preprocessing (destructive transforms) lack explicit validation checkpoints or validate→fix→retry feedback loops; per the rubric's scoring note, missing validation in batch/destructive operations caps this at 2.	2 / 3
Progressive Disclosure	Sections are organized, but six full code implementations sit inline in SKILL.md and the existing bundle file references/implementation.md is never referenced or signaled in the body, matching the anchor-2 case of content that should be separate being inline with references present but not clearly signaled.	2 / 3
	Total	9 / 12 Passed

Description

90%Weight 40%Scale 1-3

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 description with explicit what/when guidance and natural, domain-specific trigger terms. The only weakness is that the stated actions are variants of one performance goal rather than a list of distinct concrete capabilities.

Dimension	Reasoning	Score
Specificity	Quotes "Optimize Deepgram API performance for faster transcription and lower latency" and "improving transcription speed, reducing latency, or optimizing audio processing pipelines" name the domain and several facets, but the actions are largely synonymous variants of one optimization goal rather than a list of distinct concrete actions like the anchor-3 example, so it is not comprehensive enough for a 3.	2 / 3
Completeness	It explicitly states what ("Optimize Deepgram API performance for faster transcription and lower latency") and when ("Use when improving transcription speed, reducing latency, or optimizing audio processing pipelines") plus an explicit Trigger list, matching the anchor-3 example that answers both what and when with explicit triggers.	3 / 3
Trigger Term Quality	Trigger terms "speed up deepgram", "deepgram faster", "deepgram latency", "optimize transcription", and "deepgram throughput" are natural phrases a user would actually say, giving good coverage of variations; it is above level 2 because the set is varied and user-natural rather than a single keyword.	3 / 3
Distinctiveness Conflict Risk	All triggers are Deepgram-specific ("deepgram performance", "speed up deepgram", "deepgram latency") carving a clear niche unlikely to fire for non-Deepgram skills, matching the anchor-3 clear-niche case.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: jeremylongshore/claude-code-plugins-plus-skills
Path: plugins/saas-packs/deepgram-pack/skills/deepgram-performance-tuning/SKILL.md
Commit: 0540a62

Reviewed: about 18 hours 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.