deepgram-core-workflow-b
Implement real-time streaming transcription with Deepgram WebSocket. Use when building live transcription, voice interfaces, real-time captioning, or voice AI applications. Trigger: "deepgram streaming", "real-time transcription", "live transcription", "websocket transcription", "voice streaming", "deepgram live".
80
77%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/saas-packs/deepgram-pack/skills/deepgram-core-workflow-b/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 well-crafted skill description that excels in completeness, trigger term coverage, and distinctiveness. Its main weakness is that the 'what' portion could be more specific about the concrete actions performed (e.g., managing WebSocket connections, handling interim/final transcripts, configuring audio parameters). Overall it is a strong description that would perform well in skill selection.
Suggestions
Add more specific concrete actions to the capability description, e.g., 'Manages WebSocket connections, handles interim and final transcripts, configures audio encoding and language settings' to improve specificity.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (real-time streaming transcription) and a key technology (Deepgram WebSocket), but only describes one core action ('implement real-time streaming transcription') rather than listing multiple specific concrete actions like handling audio input, processing interim results, managing connection lifecycle, etc. | 2 / 3 |
Completeness | Clearly answers both 'what' (implement real-time streaming transcription with Deepgram WebSocket) and 'when' (building live transcription, voice interfaces, real-time captioning, or voice AI applications), with an explicit 'Use when' clause and explicit trigger terms. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms including 'deepgram streaming', 'real-time transcription', 'live transcription', 'websocket transcription', 'voice streaming', and 'deepgram live'. These are terms users would naturally use when seeking this functionality, with good variation coverage. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: Deepgram-specific WebSocket streaming transcription. The combination of 'Deepgram' and 'real-time/streaming' makes it very unlikely to conflict with other skills like general speech-to-text, batch transcription, or other audio processing 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 strong, highly actionable skill with excellent executable code examples covering the full spectrum of live streaming transcription use cases. Its main weaknesses are the lack of explicit validation checkpoints between steps (e.g., verifying connection before sending audio, confirming audio format) and the monolithic structure that could benefit from splitting advanced patterns into separate files. The error handling table partially compensates for missing inline validation but doesn't replace explicit workflow checkpoints.
Suggestions
Add explicit validation checkpoints: after Step 1, include a verification step to confirm the WebSocket connection is open and receiving data before proceeding to microphone capture.
Split advanced patterns (Steps 4-6: auto-reconnect, SSE endpoint, KeepAlive) into a separate ADVANCED.md file, keeping SKILL.md focused on the core workflow (Steps 1-3) with clear references.
Add a quick diagnostic feedback loop: e.g., 'If no transcripts appear within 5 seconds, verify audio format by logging chunk sizes and checking getReadyState() === 1'.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient with executable code examples, but includes some unnecessary commentary (e.g., the Overview section restates what the steps already show, and inline comments sometimes explain obvious things like '// Quiet (no progress)'). The error handling table and output section add some redundancy but are borderline useful. | 2 / 3 |
Actionability | Every step provides fully executable TypeScript code with specific SDK methods, configuration options, and concrete patterns. Code is copy-paste ready with real API calls, proper event handling, and practical patterns like auto-reconnect with backoff and SSE endpoints. | 3 / 3 |
Workflow Clarity | Steps are clearly sequenced and build progressively from basic connection to advanced patterns. However, there are no explicit validation checkpoints — for example, no step to verify the WebSocket connection is working before sending audio, no verification that audio format matches configuration, and no feedback loop for debugging when transcripts aren't received. | 2 / 3 |
Progressive Disclosure | The content is well-structured with clear sections, but it's quite long (~200 lines of code) with all content inline. The diarization, auto-reconnect, SSE endpoint, and KeepAlive sections could be split into separate reference files. External links to Deepgram docs are provided but no bundle files exist to offload advanced patterns. | 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.