CtrlK
BlogDocsLog inGet started
Tessl Logo

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".

72

Quality

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Quality

Content

92%

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

A highly actionable, lean, well-sequenced body with strong executable examples and clear error guidance. Its main weakness is progressive disclosure: an existing reference file is never linked, so the skill reads as a monolithic code dump rather than an overview pointing to deeper material.

Suggestions

Link references/streaming-implementation.md from the body (e.g., a '## Advanced implementation' section: 'See [streaming-implementation.md](references/streaming-implementation.md)') so the provided bundle file is actually signaled and used.

Move the longer worked examples (e.g. the SSE endpoint or auto-reconnect class) into the reference and keep SKILL.md as a concise overview with one-level-deep pointers.

Make workflow checkpoints explicit where state transitions matter (e.g. 'verify connection.getReadyState() === 1 before send' as a stated step rather than only an inline guard).

DimensionReasoningScore

Conciseness

The body is lean executable code with minimal prose; inline comments are brief and serve correctness (e.g. "interim_results: true // Show in-progress results") rather than padding, and it does not explain concepts Claude already knows, matching the level-3 anchor; it is not level 2 because there is no excess explanatory fluff to tighten.

3 / 3

Actionability

All six steps are complete, copy-paste-ready TypeScript with concrete parameters and commands (Sox flags, SDK methods, SSE headers), matching the level-3 anchor; it is not level 2 because the examples are fully executable rather than pseudocode or partial.

3 / 3

Workflow Clarity

Steps are clearly sequenced from basic transcription through mic capture, diarization, reconnect, SSE, and keepalive, with an error-handling table and a backoff/retry loop; the operations are non-destructive streaming so the destructive-batch cap does not apply, matching the level-3 anchor. It sits just above level 2 because checkpoints are implicit rather than explicit validate-then-proceed gates.

3 / 3

Progressive Disclosure

A bundle file (references/streaming-implementation.md) exists but is never referenced or signaled from the body, while six substantial code blocks are inlined monolithically, matching the level-2 anchor ('content that should be separate is inline' / 'references present but not clearly signaled'); it is not level 3 because the available reference is not used to offload detail, and not level 1 because sections are well organized and references are only one level deep.

2 / 3

Total

11

/

12

Passed

Description

90%

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, well-targeted description that covers what and when with explicit triggers in third person and a distinct Deepgram niche. The only weakness is that the 'what' centers on a single overarching action rather than enumerating several concrete capabilities.

DimensionReasoningScore

Specificity

It names the domain ("Deepgram WebSocket") and a concrete action ("Implement real-time streaming transcription") but describes a single main capability rather than listing multiple distinct concrete actions, matching the level-2 anchor; it falls short of level 3's 'lists multiple specific concrete actions' because the verb set is essentially one (streaming transcription).

2 / 3

Completeness

It answers both 'what' ("Implement real-time streaming transcription with Deepgram WebSocket") and 'when' ("Use when building live transcription, voice interfaces, real-time captioning, or voice AI applications") with an explicit 'Use when' clause and triggers, matching the level-3 anchor; voice is third person ("Implement") with no first/second person, so no penalty applies.

3 / 3

Trigger Term Quality

The explicit Trigger list ("deepgram streaming", "real-time transcription", "live transcription", "websocket transcription", "voice streaming", "deepgram live") gives good coverage of natural phrasings a user would actually say, matching the level-3 anchor; it is not level 2 because common variations are well covered rather than partial.

3 / 3

Distinctiveness Conflict Risk

The Deepgram-specific niche plus branded trigger terms make it clearly distinguishable and unlikely to fire for unrelated skills, matching the level-3 anchor; it is not level 2 because the vendor lock-in to Deepgram and WebSocket streaming sharply limits overlap.

3 / 3

Total

11

/

12

Passed

Validation

87%

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

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

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
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.