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".
80
77%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/saas-packs/deepgram-pack/skills/deepgram-performance-tuning/SKILL.mdQuality
Discovery
89%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 a solid description with excellent trigger terms and clear 'when' guidance, making it highly distinguishable and complete. Its main weakness is that the 'what' portion is somewhat general—it says 'optimize performance' without listing specific concrete actions (e.g., configuring streaming endpoints, adjusting encoding parameters, implementing connection pooling). Adding 2-3 specific optimization techniques would strengthen it.
Suggestions
Add specific concrete actions to the description, e.g., 'configure streaming endpoints, adjust encoding parameters, implement connection pooling, tune batch sizes' to improve specificity.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (Deepgram API performance) and a general action (optimize for faster transcription and lower latency), but does not list multiple specific concrete actions like caching strategies, batch processing, endpoint configuration, or streaming setup. | 2 / 3 |
Completeness | Clearly answers both 'what' (optimize Deepgram API performance for faster transcription and lower latency) and 'when' (explicit 'Use when' clause plus a 'Trigger' list with specific phrases). | 3 / 3 |
Trigger Term Quality | Includes a well-curated set of natural trigger terms users would actually say: 'deepgram performance', 'speed up deepgram', 'optimize transcription', 'deepgram latency', 'deepgram faster', 'deepgram throughput'. These cover common variations of how users would phrase their needs. | 3 / 3 |
Distinctiveness Conflict Risk | Highly specific to Deepgram API performance optimization, which is a clear niche. The trigger terms are all Deepgram-specific, making it very unlikely to conflict with general transcription or other API skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
64%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, highly actionable skill with executable code for every optimization lever. Its main weaknesses are the lack of integrated validation checkpoints in the workflow (especially for audio preprocessing and batch operations) and the monolithic structure that could benefit from splitting detailed implementations into separate files. The performance levers summary table at the top is an excellent quick-reference pattern.
Suggestions
Add explicit validation checkpoints after audio preprocessing (e.g., verify output file exists, check duration/format with ffprobe) and after batch processing (e.g., check fulfilled vs rejected ratio before proceeding).
Consider splitting the detailed code implementations (Steps 3-6) into a separate EXAMPLES.md or RECIPES.md file, keeping SKILL.md focused on the decision framework (the performance levers table) and minimal quick-start code for Steps 1-2.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly efficient with its table summaries and code examples, but includes some unnecessary commentary (e.g., explaining what Nova-3 vs Nova-2 is, the console.log statements for savings reporting). The code blocks are lengthy and could be trimmed while preserving clarity. The overview section is well-condensed though. | 2 / 3 |
Actionability | Every step includes fully executable, copy-paste ready code in both bash and TypeScript. The ffmpeg commands are complete with flags explained inline, the TypeScript functions are typed and importable, and specific library usage (p-limit, ioredis, @deepgram/sdk) is shown with real API calls. | 3 / 3 |
Workflow Clarity | Steps are clearly sequenced (preprocess → model selection → streaming → batch → caching → benchmarking), but there are no explicit validation checkpoints between steps. For a pipeline involving audio preprocessing and batch operations, there should be verification steps (e.g., validate ffmpeg output before sending to API, verify preprocessed audio quality). The error handling table at the end is helpful but reactive rather than integrated into the workflow. | 2 / 3 |
Progressive Disclosure | The content is well-structured with clear sections and a summary table up front, but it's a monolithic ~200-line file with no bundle files to offload detailed code examples. The streaming, caching, and benchmarking sections could be separate reference files, with SKILL.md providing just the overview table and quick-start patterns. External resource links at the end are good but the inline content is heavy. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 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 | 9 / 11 Passed | |
- Commit
3a2d27d
- 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.