CtrlK
BlogDocsLog inGet started
Tessl Logo

tts-voice-synthesis

智能语音合成服务,支持音色克隆、拟人化语义适配配音、流式实时生成、多语言与方言支持,提供 1.7B/0.6B 双模型选择

64

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/tts-voice-synthesis/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

72%

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

The content is highly actionable with copy-paste-ready commands and well-structured progressive disclosure to real reference files, but it is somewhat verbose with agent-behavior narration and lacks rigorous validate→fix→retry feedback loops for streaming and cloning operations. Tightening prose and adding explicit validation checkpoints would lift the weaker dimensions.

Suggestions

Remove narration of obvious agent behavior (e.g., '智能体将分析文本情绪', '确保分段自然,不会截断语义') to tighten conciseness toward score 3.

Add explicit validate→fix→retry feedback loops for voice cloning and streaming (e.g., verify the cloned voice on a sample sentence and re-extract if quality is poor; validate merged audio length against input text before finalizing).

De-duplicate the inline 使用示例 examples with references/usage_examples.md, keeping the body examples minimal and pointing to the reference for the full set.

DimensionReasoningScore

Conciseness

The body avoids teaching known concepts, but it is padded with narration of obvious agent behavior ('智能体将分析文本情绪和语义特征', '确保分段自然,不会截断语义') and soft filler ('如有需要,调整参数重新生成'), so it could be tightened rather than being lean throughout. It is not score 1 (no concept-teaching fluff) and not score 3 (agent-narration padding present).

2 / 3

Actionability

Provides fully executable bash commands with concrete flags (--text, --output_path, --model_size, --voice, --emotion, --speed, --pitch, --streaming, --reference_audio) pointing to real, verified scripts, making the examples copy-paste ready across all four modes. Matches the score-3 anchor for executable, specific, copy-paste-ready guidance.

3 / 3

Workflow Clarity

Each of the four modes is sequenced with numbered steps and includes a verification step, but the checkpoints are soft ('如有需要,调整参数重新生成') with no explicit validate→fix→retry loop, and batch/streaming operations (which produce '多个临时音频片段') lack a hard validation checkpoint, capping the score at 2. Steps are listed with implicit validation gaps; not score 3 because feedback loops are missing for batch-like streaming/cloning operations.

2 / 3

Progressive Disclosure

The body is a clear overview that signals one-level-deep references to verified files (references/model_config.md, references/emotion_guide.md, references/usage_examples.md) plus scripts, with a 资源索引 navigation section and no deep nesting, matching the score-3 anchor. Slight inline duplication with usage_examples.md, but structure and navigation are otherwise clean.

3 / 3

Total

10

/

12

Passed

Description

82%

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 and distinctive, naming five concrete capabilities and using natural trigger terms, but it omits any explicit 'Use when' guidance, which caps completeness at 2. Adding an explicit trigger clause would raise the completeness and overall score.

Suggestions

Append an explicit 'Use when...' / '当需要...时使用' trigger clause to the description (e.g., '当需要将文本转为语音、克隆音色或生成情感化配音时使用') so completeness can score 3.

Add common layperson phrasings like '文字转语音' or 'TTS' alongside '语音合成' to broaden trigger-term coverage for users who don't use the technical term.

Keep the existing concrete capability list as-is; it is already specific and distinct, so no structural change is needed beyond the trigger clause.

DimensionReasoningScore

Specificity

Lists multiple concrete capabilities explicitly ('支持音色克隆', '拟人化语义适配配音', '流式实时生成', '多语言与方言支持', '提供 1.7B/0.6B 双模型选择'), matching the score-3 anchor for several specific actions rather than just naming a domain.

3 / 3

Completeness

Clearly answers 'what' (a voice-synthesis service with cloning, emotion dubbing, streaming, multilingual support), but the description itself lacks any explicit 'Use when...' / '当...时使用' trigger, so per the guideline completeness is capped at 2.

2 / 3

Trigger Term Quality

Natural domain terms a user would say are present ('语音合成', '配音', '音色克隆', '流式'), giving good coverage of the terms users actually voice when needing TTS; not score 2 because the keywords are natural rather than technical jargon or generic.

3 / 3

Distinctiveness Conflict Risk

The niche is specific (TTS with cloning, emotion adaptation, streaming) with distinctive triggers unlikely to fire for unrelated skills, matching the score-3 anchor for a clear niche with distinct triggers.

3 / 3

Total

11

/

12

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
anbeime/skill
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.