speech-to-text

Transcribe audio to text using ElevenLabs Scribe v2. Use when converting audio/video to text, generating subtitles, transcribing meetings, or processing spoken content.

1.88x

Quality

92%

Does it follow best practices?

Impact

96%

1.88x

Average score across 11 eval scenarios

Securityby

Passed

No findings from the security scan

Quality

Content

85%

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

The body is highly actionable with executable examples across languages and well-organized progressive disclosure into real reference files. Its main weakness is mild verbosity in a few explanatory passages that restate use cases already covered by the description.

Suggestions

Trim explanatory padding that restates use cases already in the description, e.g. remove "useful for meetings, interviews, or any multi-speaker audio" from the Speaker Diarization section and tighten the commit definition.

Move the full Event Types and Word types reference tables into the existing realtime-events.md or transcription-options.md references, keeping only a one-line pointer in SKILL.md to reduce inline reference bulk.

Consolidate the three Quick Start language examples into a single primary example plus a brief note linking to language-specific details, since the per-language variants inflate token cost without adding distinct capability guidance.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean code-plus-one-line explanations and avoids teaching basic concepts, but contains some unnecessary explanatory padding such as "useful for meetings, interviews, or any multi-speaker audio" and "A 'commit' tells the model to finalize the current segment" that could be tightened, matching the mostly-efficient score-2 anchor rather than the every-token-earns-its-place score-3 case.	2 / 3
Actionability	Executable Python, JavaScript, cURL, and TypeScript snippets with concrete parameters (model_id, diarize, keyterms) are copy-paste ready and complete, matching the score-3 anchor rather than the pseudocode/incomplete score-2 case.	3 / 3
Workflow Clarity	The core single task is unambiguous via Quick Start, and the real-time section sequences connect-stream-handle-events clearly; no destructive or batch operations require validation feedback loops, so per the simple-skills note the score-3 anchor applies.	3 / 3
Progressive Disclosure	A concise overview with clearly signaled one-level-deep references to six verified real files (installation.md, transcription-options.md, realtime-*.md) under a References section matches the score-3 anchor; no deep nesting or monolithic inline reference material.	3 / 3
	Total	11 / 12 Passed

Description

100%

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 is concise, specific, and uses third-person imperative voice while clearly stating both capabilities and explicit trigger conditions. It avoids vague fluff and over-claims, covering natural user language well.

Dimension	Reasoning	Score
Specificity	"Transcribe audio to text using ElevenLabs Scribe v2" names a concrete action and tool, and the use-case list (converting audio/video to text, generating subtitles, transcribing meetings) enumerates multiple specific concrete actions, matching the score-3 anchor rather than the partial coverage at score 2.	3 / 3
Completeness	It answers both what ("Transcribe audio to text using ElevenLabs Scribe v2") and when ("Use when converting audio/video to text, generating subtitles..."), with an explicit trigger clause matching the score-3 anchor and avoiding the score-2 cap for a missing 'Use when' clause.	3 / 3
Trigger Term Quality	Phrases like "converting audio/video to text, generating subtitles, transcribing meetings, or processing spoken content" cover natural terms a user would say, matching the good-coverage score-3 anchor rather than the narrower score-2 case.	3 / 3
Distinctiveness Conflict Risk	The ElevenLabs Scribe v2 niche with speech-to-text-specific triggers (subtitles, meetings, spoken content) is clearly distinguishable and unlikely to trigger the wrong skill, matching the score-3 anchor rather than the overlapping score-2 case.	3 / 3
	Total	12 / 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
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: calesthio/OpenMontage
Path: .agents/skills/speech-to-text/SKILL.md
Commit: 0af32ce

Reviewed: 1 day 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.