القائمة
Deep pipeline diagnostic — every model, system specs, audio config, speaker state. Flags missing or broken components. Use for diagnose, debug, or health check.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Collaborate over x0x — contacts, messages, spaces, boards, files, presence, trusted-machine port forwards, and replicated stores. Use for sharing cards, connecting people, or cross-device work.
Guide for surfacing one relevant Fae capability that the user hasn't set up yet. Warm, specific, one thing at a time.
Extract training signals from Fae's memory — SFT examples, DPO correction pairs, engagement scores, and interest-weighted sampling.
Set up (or turn off) a cloud brain for harder questions via OpenRouter — privacy-first, plain language. Use when the user wants a bigger or smarter brain or mentions the cloud.
A warm first conversation for someone meeting Fae for the first time — getting to know their name, where they live, and what they care about, showing one thing live, and explaining how to talk to her.
Phase C headless proof skill — prints a deterministic marker via uv run --script through the governed ToolHost bash path.
| name | self-diagnostic |
| description | Deep pipeline diagnostic — every model, system specs, audio config, speaker state. Flags missing or broken components. Use for diagnose, debug, or health check. |
| metadata | {"author":"fae","version":"2.0","tags":["system","diagnostics","debug"]} |
You are running a comprehensive self-diagnostic. Work through EVERY section methodically. Report findings clearly in spoken language. Be specific about what IS loaded, what's MISSING, and what's BROKEN.
Use bash to gather machine specs:
bash "sysctl -n hw.model hw.memsize machdep.cpu.brand_string 2>/dev/null; echo '---'; system_profiler SPHardwareDataType 2>/dev/null | grep -E 'Model|Chip|Memory|Cores'"
Report:
bash "top -l 1 -n 0 | head -12; echo '---'; vm_stat | head -5"
Report: Current memory pressure, CPU load, swap usage. Flag if memory pressure is critical.
bash "df -h / | tail -1"
Report: Disk space. Flag if below 10 GB free (model downloads need space).
This is the most important part. List EVERY model in the pipeline with its load status.
~/Library/Caches/fae/ by MLX on first launchfae_self echo rejection only — prevents Fae from hearing her own TTS output (voice identity retired in S18; PTT is the access gate)wespeaker.mlmodelc in Resources/Models/SpeakerEncoder/bash "ls -la ~/Library/Application\\ Support/fae/speakers.json 2>/dev/null | head -1 || echo 'No speaker profiles file'"
Check and report:
bash "system_profiler SPAudioDataType 2>/dev/null | head -20"
Report:
Check and report:
bash "ls -la ~/Library/Application\\ Support/fae/speakers.json 2>/dev/null || echo 'No speaker profiles file'"
Report from the JSON (if it exists):
scheduler_list
Report:
Report from context:
bash "tail -10 ~/Library/Application\\ Support/fae/security-events.jsonl 2>/dev/null || echo 'No security log'"
Report: Recent security events. Flag any errors.
Check macOS permissions:
Structure your spoken summary as:
"Here's my full diagnostic report:"
| Issue | Fix |
|---|---|
| Mel-spectral fallback | "WeSpeaker model not loading — update to latest version or reinstall" |
| fae_self profile missing | "Echo-rejection profile not found — restart Fae to rebuild it" |
| Low disk space | "Free up disk space — I need room for model downloads" |
| Voice Isolation off | "Switch to Voice Isolation in Control Center for cleaner audio" |
| VP disabled (dev mode) | "You're in dev mode — Voice Processing is disabled for testing" |
| Stale scheduler tasks | "Some background tasks haven't run — try restarting me" |
End with: "That's the full picture. Let me know if you want me to dig deeper into anything."