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

77%

Does it follow best practices?

Impact

96%

1.88x

Average score across 11 eval scenarios

Securityby

—

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/speech-to-text/SKILL.md

Quality

Content

64%Scale 1-3

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, actionable skill with excellent executable code examples across multiple languages and good coverage of the ElevenLabs STT API surface. Its main weaknesses are that it's too long for a SKILL.md overview — particularly the real-time streaming section which duplicates content that should live in the referenced files — and it includes some explanatory prose that Claude doesn't need.

Suggestions

Move the bulk of the real-time streaming section (React examples, commit strategies, event types) into the referenced files, keeping only a minimal quick-start example in the main skill

Remove explanatory phrases like 'Identify WHO said WHAT', 'Help the model recognize specific words it might otherwise mishear', and the word type descriptions — Claude can infer these from the code and response format

Add a brief validation step or note about checking transcription quality (e.g., checking language_probability threshold) to improve workflow clarity for production use cases

Dimension	Reasoning	Score
Conciseness	The skill is mostly efficient with good code examples, but includes some unnecessary explanations Claude already knows (e.g., explaining what speaker diarization is, what word types mean, what partial vs committed transcripts are). The real-time section is quite lengthy and could be trimmed or moved to references.	2 / 3
Actionability	Excellent executable code examples across Python, JavaScript, and cURL. Every feature (timestamps, diarization, keyterms, language detection, real-time streaming) has copy-paste ready code with specific parameters and expected outputs.	3 / 3
Workflow Clarity	The skill covers individual API calls well but lacks a clear multi-step workflow for common use cases (e.g., transcribe → validate output → handle errors → retry). Error handling is present but basic, and there's no feedback loop for verifying transcription quality or handling partial failures on large files.	2 / 3
Progressive Disclosure	References are well-organized at the bottom with clear one-level-deep links, but too much detail is inline — the real-time streaming section (~60 lines with React examples, commit strategies, and event types) should largely be in the referenced files rather than in the main skill. The main file tries to be both overview and detailed reference.	2 / 3
	Total	9 / 12 Passed

Description

89%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.

This is a well-structured skill description that clearly communicates its purpose and when to use it. The explicit 'Use when...' clause with multiple trigger scenarios is a strong point. The main weakness is that the 'what' portion could list more specific capabilities beyond the single action of transcribing audio to text.

Suggestions

Expand the capabilities list to include specific features like speaker diarization, timestamp generation, multi-language support, or supported file formats to improve specificity.

Dimension	Reasoning	Score
Specificity	Names the domain (audio transcription) and the specific tool (ElevenLabs Scribe v2), but only describes one core action ('transcribe audio to text') rather than listing multiple concrete capabilities like speaker diarization, timestamp generation, or language detection.	2 / 3
Completeness	Clearly answers both 'what' (transcribe audio to text using ElevenLabs Scribe v2) and 'when' with an explicit 'Use when...' clause covering multiple trigger scenarios (converting audio/video, generating subtitles, transcribing meetings, processing spoken content).	3 / 3
Trigger Term Quality	Includes strong natural trigger terms users would say: 'audio', 'video', 'text', 'subtitles', 'transcribing', 'meetings', 'spoken content'. These cover a good range of common user phrasings when needing transcription.	3 / 3
Distinctiveness Conflict Risk	Highly distinctive due to the specific tool name (ElevenLabs Scribe v2) and the clear niche of audio-to-text transcription. Unlikely to conflict with other skills unless there are multiple transcription tools available.	3 / 3
	Total	11 / 12 Passed

Validation

81%

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 — 9 / 11 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	9 / 11 Passed

Repository: calesthio/OpenMontage
Path: .agents/skills/speech-to-text/SKILL.md
Commit: 08e2151

Reviewed: 17 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.