heymegabyte/agent-skills / 07-quality-and-verification
quality-and-verification
Unverified5-level verification pyramid: static→unit→Playwright E2E (homepage-first, 6bp)→AI visual→post-deploy. 8-check quality gate. Multi-agent testing (functional/security/a11y/performance). Playwright v1.56+ AI agents (Planner/Generator/Healer, v1.63 current). WCAG 2.2 AA via axe-core v4.13. Percy+Chromatic visual regression. ADA Title II 2027/2028 deadlines.
picovoice-eagle-biometric.md
View on GitHub---
name: "Picovoice Eagle Voice Biometric"
description: "Tier-2 voice biometric using @picovoice/eagle-node. On-device, zero-network, free tier. Sits between ONNX ECAPA-TDNN (tier-1) and MFCC cosine (tier-3) in the verification chain."
updated: "2026-04-23"
---
# Picovoice Eagle Voice Biometric Pattern
## Integration
- `@picovoice/eagle-node` v1.2+ as `optionalDependency`
- Lazy-loaded on first use — never imported if ONNX model loads successfully
- Requires `PICOVOICE_ACCESS_KEY` env var (free tier: 3 voice profiles)
## Three-Tier Chain
1. **ONNX ECAPA-TDNN** — 192-dim embedding, best accuracy
2. **Picovoice Eagle** — on-device, free tier, no model download
3. **MFCC cosine similarity** — zero dependencies, zero cost
`loadModel()` tries each tier in order, first success wins. Response includes `engine` field (`'ecapa-tdnn' | 'eagle' | 'mfcc'`).
## Eagle Enrollment
- `EagleProfiler` processes Int16 PCM frames at `profiler.frameLength` samples
- Feed frames until `result.percentage >= 100` (minimum 50% for usable profile)
- Export profile bytes → store as Float32Array embedding
- `profiler.reset()` after each enrollment
## Eagle Verification
- `Eagle` instance created with access key + stored profile
- `eagle.process(frame)` returns score 0.0-1.0
- Threshold: 0.5
- Process all frames, take max score
- Cleanup: `eagle.release()` after verification
## Audio Format
- **Input:** Float32 PCM 16kHz
- **Convert to Int16:** `s < 0 ? s * 0x8000 : s * 0x7FFF`
- **Frame size:** from `eagleProfiler.frameLength`
- If enrollment percentage <50%, falls back to MFCC fingerprint extraction
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |