AB characteristic-voice
Use this skill whenever the user wants speech to sound more human, companion-like, or emotionally expressive. Triggers include: any mention of 'say like', 'talk like', 'speak like', 'companion voice', 'comfort me', 'cheer me up', 'sound more human', 'good night voice', 'good morning voice', or requests to add fillers, emotion, or personality to generated speech. Also use when the user wants to mimic a specific character's voice, apply speaking style presets (goodnight, morning, comfort, celebration, chatting), tune emotional parameters like warmth or tenderness, or make TTS output feel like a real person talking. If the user asks for a 'voice message', 'companion audio', 'character voice', or wants speech that sighs, laughs, hesitates, or sounds genuinely warm, use this skill. Do NOT use for plain text-to-speech without personality, music generation, sound effects, or general coding tasks unrelated to expressive speech.
As a process B 65/100 · Nearly there — weak spots: result and completion, consistency, progress reporting
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 1
✓ No critical or high findings
Medium and low: 1
-
low Exfiltration
net-credential-usescripts/speak.sh:175Credential used in a network call (verify the destination is the intended service) (quoted — discussed, not commanded)resp="$(curl -sS -H "Authorization: ${api_key}" \quoted
Files scanned: 3. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 65/100
- 0Result and completion. Does not say what the result is
- 0Progress reporting. Says nothing while it works
- 40Consistency. Frontmatter name (characteristic-voice) differs from the folder (noizai-characteristic-voice)
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 65Failures and branches. 3 branches
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Steps. 20 steps
- 100When it triggers. States when to use and when not to
- 100Execution cost. Instruction body is 1554 tokens
- 100Running it twice. Mutating operations check current state
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 10 top-level sections: this looks like several domains in one skill
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +3Description length 933: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +4Description says when NOT to use the skill
- +4Structure: 18 headings
- +3Step-by-step instructions: 20 items
- +4Has examples (4 code blocks)
- +3All 1 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 88.