Skill audit report
voice-likeness-eval Audit report.
Measure how much a cloned voice sounds like the real character, and whether a change really helped. Covers speaker-embedding similarity (ECAPA) against the centroid of held-out real clips, word error rate with Whisper, comparing per line with paired t-tests, how many runs it takes to beat sampling noise, checking a take-picker with an independent speaker model, and simulating picking strategies on many scored takes. Use when comparing voice engines or settings (Fish Audio Drama 3 vs S2 Pro vs ElevenLabs, sampling, reference clips, best-of-N, fine-tuning), or before telling anyone that one voice setup is better than another.
OpenAgentSkill Trust Score
OpenAgentSkill Trust Score
The Trust Score helps an agent decide whether a skill is safe enough to shortlist before installation.
GitHub adoption
WARN48
54 GitHub stars
Stars/forks activity
WARN43
54 stars, 4 forks; issue activity unavailable in current metadata
Recent maintenance
PASS100
5d since push
License clarity
PASS86
MIT
README/SKILL.md completeness
INFO76
Public metadata needs stronger README/SKILL.md context
Dependency/runtime risk
INFO72
command execution surface
Install availability
PASS92
npx skills add christos-antono/NPCTalk --skill voice-likeness-eval
Install command safety
INFO68
dynamic command execution, standard package or runtime install path
Permission surface
INFO62
shell or command execution, filesystem or document access
Repository evidence
PASS86
https://github.com/christos-antono/NPCTalk/tree/main/.claude/skills/voice-likeness-eval
Review status
WARN46
AI review approval is missing
Agent Proven outcomes
INFO54
No agent outcome data yet
Checks
Install and adoption review
Install path
92
npx skills add christos-antono/NPCTalk --skill voice-likeness-eval
Repository
88
https://github.com/christos-antono/NPCTalk/tree/main/.claude/skills/voice-likeness-eval
License
86
MIT
Maintenance
100
5d since push
AI review
55
Review approval is missing
README/SKILL.md completeness
84
Usable description available
Dependency risk
72
command execution surface
Install command safety
68
dynamic command execution, standard package or runtime install path
Permission surface
62
shell or command execution, filesystem or document access
Stars/forks activity
43
54 stars, 4 forks; issue activity unavailable in current metadata
Adoption
68
54 GitHub stars
Warnings
- AI review approval is missing
- Quality score needs review
- GitHub adoption: 54 GitHub stars
- Stars/forks activity: 54 stars, 4 forks; issue activity unavailable in current metadata
- Review status: AI review approval is missing
Method
This report combines public metadata, AI review output, repository freshness, install readiness, OpenAgentSkill events, quality scoring, trust checks, and the agent safety gate. It is not a full source-code security review.
Compare nearby options
Related skills to audit next
openai-whisper
Local speech-to-text with the Whisper CLI (no API key).
391K Stars · Audit report
events
When the user wants to plan, run, sponsor, speak at, or get pipeline from events — webinars, conferences, trade shows, meetups, dinners, workshops, virtual summits, or user conferences. Also use when the user mentions 'event marketing,' 'field marketing,' 'run a webinar,' 'webinar funnel,' 'show-up rate,' 'should we sponsor,' 'sponsor a conference,' 'trade show booth,' 'booth strategy,' 'event ROI,' 'badge scans,' 'event follow-up,' 'speaking slot,' 'CFP,' 'conference talk,' 'host a dinner,' 'user conference,' or 'virtual summit.' Covers all four roles: hosting, sponsoring/exhibiting, speaking, and attending. For product launch moments, see launch. For the partnership side of joint webinars, see co-marketing. For ongoing community programs, see community-marketing. For podcast appearances, see public-relations. For the email sequences themselves, see emails.
55K Stars · Audit report
nemo-curator
GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16× faster), quality filtering (30+ heuristics), semantic deduplication, PII redaction, NSFW detection. Scales across GPUs with RAPIDS. Use for preparing high-quality training datasets, cleaning web data, or deduplicating large corpora.
13K Stars · Audit report