text-to-speech

Generate speech audio from text using HeyGen's Starfish TTS model. Use when: (1) Generating standalone speech audio files from text, (2) Converting text to speech with voice selection, speed, and pitch control, (3) Creating audio for voiceovers, narration, or podcasts, (4) Working with HeyGen's /v1/audio endpoints, (5) Listing available TTS voices by language or gender.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

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

Quality

Content

71%Weight 40%Scale 1-5

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 complete executable examples and a clear default workflow, but it is token-heavy due to triply-duplicated language examples and inlined response schemas, and it would benefit from moving detailed API reference into reference files. Workflow clarity is strong but lacks an explicit post-generation verification checkpoint.

Suggestions

Move the full per-language (curl/TypeScript/Python) implementations and response-format JSON into a references/ file, keeping only one compact example plus the field tables inline in SKILL.md to improve conciseness and progressive disclosure.

Add an explicit verification step to the Default Workflow, e.g. 'After generation, fetch the audio_url and confirm duration matches expectations; re-generate if the sample sounds flat or ignores break tags'.

Trim redundant explanation (e.g. the same voices endpoint is shown via curl twice and as full TS/Python) to reduce tokens while preserving the MCP-preferred guidance.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with concrete reference tables and code, but the full curl/TypeScript/Python examples are repeated for both endpoints and response-format JSON blocks are inlined wholesale, adding tokens a Claude-capable reader could partly do without.	3 / 5
Actionability	Fully executable copy-paste-ready curl, TypeScript, and Python snippets for both endpoints, plus field tables, real response shapes, and worked end-to-end examples covering the common cases.	5 / 5
Workflow Clarity	A clear numbered Default Workflow sequences voice-list -> pick -> generate -> use audio_url, and Best Practices add validation cues (check support_locale, preview voice), but there is no explicit validate/verify checkpoint before or after generation for batch/destructive-style runs.	4 / 5
Progressive Disclosure	Content is reasonably section-organized, but the large duplicated per-language code blocks and full response-format JSON are inlined in SKILL.md with no bundle files to offload detail; with no references/ or assets/ present there is no one-level-deep navigation.	3 / 5
	Total	15 / 20 Passed

Description

87%Weight 40%Scale 1-5

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 specific, well-triggered, and explicitly answers both what and when, with a clear HeyGen Starfish TTS niche that minimizes conflict with other skills. It is among the stronger descriptions, with only minor room for additional natural synonyms.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('Generating standalone speech audio files', 'Converting text to speech with voice selection, speed, and pitch control', 'Listing available TTS voices by language or gender') plus concrete API endpoints, with minor gaps (e.g. pitch control only implied via 'pitch control').	4 / 5
Completeness	Clearly states what it does ('Generate speech audio from text using HeyGen's Starfish TTS model') and explicitly gives 'Use when:' with five concrete trigger cases, fully answering both what and when.	5 / 5
Trigger Term Quality	Strong natural coverage ('speech audio', 'text to speech', 'voiceovers', 'narration', 'podcasts', 'TTS voices', 'voice selection'), with a few natural variants missing (e.g. 'read aloud', 'audio narration', '.mp3').	4 / 5
Distinctiveness Conflict Risk	Scoped tightly to HeyGen's Starfish TTS and /v1/audio endpoints with explicit standalone-audio boundary, making it clearly distinct from HeyGen video skills and minimal conflict risk.	5 / 5
	Total	18 / 20 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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	13 / 16 Passed

Repository: calesthio/OpenMontage
Path: .agents/skills/text-to-speech/SKILL.md
Commit: 1bab711

Reviewed: about 20 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.