122 tagged Apple Silicon, measured the same way as everything else here.
Browse within: mlx 23desktop-app 18ai-chat 17chatgpt 17deepseek 17asr 16macos 16speech-to-text 16coreml 15agent-inventory 12agent-observability 12ai-agent-monitoring 12bun 12openclaw 12
Skill Claude CodeCodex
Use this skill when the user wants to add or port a new model architecture to MLX-VLM — mapping a Hugging Face modeltype to a new file under mlxvlm/models, writing the ModelConfig, matching layer/weight names, reusing a similar existing model, adding a test class, and validating the port. Covers vision-language…
Skill Claude CodeCodex
Use this skill when the user wants to benchmark an MLX-VLM change and present the numbers in a PR — fork-vs-main A/B comparisons, isolated-module micro-benchmarks, median-of-N timing with warmup, peak-memory reporting, correctness checks, parameter sweeps, and self-contained reproducible bench scripts to paste into a…
Skill Claude CodeCodex
Use this skill when the user wants to convert a Hugging Face model to MLX or quantize/dequantize one with mlxvlm.convert, including bits and group size, quant modes (affine, mxfp4, nvfp4, mxfp8), RTN vs AWQ, mixed-bit recipes, dtype casts, calibration (text or multimodal), local vs Hub paths, revisions, uploading to…
Skill Claude CodeCodex
Autonomous performance optimization: research, PoC, benchmark, implement, review, PR.
Skill Claude CodeCodexCursor
Debug MLX ports by comparing against a PyTorch/diffusers reference via exported tensors/images (export-then-compare).
Skill Claude CodeCodexCursor
Port ML models into mflux/MLX with correctness-first validation, then refactor toward mflux style.
Skill Claude CodeCodexCursor
Write a hermetic "tiny" model-saving test for an mflux model — a fast twin of the slow save/load test that runs the real ModelSaver/WeightLoader/WeightApplier seam on real component classes at toy dimensions, with no downloads or image generation. Use when asked to "make tiny test for ".
Skill Claude CodeCodex
Read macOS Calendar events via the icalBuddy CLI and create events via AppleScript (osascript). Use to check the user's calendar, agenda, upcoming events, or add an event on macOS.
Skill Claude CodeCodex
Manage Apple Reminders via the remindctl CLI on macOS — list, add, complete, delete, manage lists.
Skill Claude CodeCodex
Use the gog CLI for Google Workspace tasks across Gmail, Calendar, Drive, Docs, Sheets, Contacts, and related services. Use when the user asks to check email, search Gmail, inspect calendar events, find Drive files, read Docs or Sheets, or manage Google Workspace data through gog.
Skill Claude CodeCodex
Sync Granola meeting notes into Steno's file-based store. Reads recent meetings from the Granola MCP (titles, dates, participants, AI summaries, verbatim transcripts) and regenerates Steno's output/summary.md and transcripts/transcript.txt files in a target folder (e.g. an iCloud Drive directory). The operation is…
Skill Claude CodeCodex
Make the agent aware of the user's Steno meeting notes and able to pull them in on request. Steno (https://github.com/stenolabs/stenoai) records, transcribes, and summarizes meetings entirely on-device, storing each as a Markdown summary plus a transcript. This skill locates that store cross-platform (macOS, Windows…
Skill Claude CodeCodex
Run performance + quality benchmarks. ASR reports WER, RTF, process memory, and throughput across engines/variants. Arguments include asr, tts, vad, diarize, asr-quick.
Skill Claude CodeCodex
Review a pull request for conceptual fit, architecture impact, adversarial failure modes, security risk, docs impact, regression risk, test coverage, and merge readiness. Use when asked to review a PR, check whether a PR is safe to merge, decide if more tests are needed, perform adversarial or security review, or…
Skill Claude CodeCodex
Run tests. Use after code changes to validate. Arguments: unit (default, no GPU), e2e (with models), filter name, or all.
Skill Claude CodeCodex
Skill "bench" from ddalcu/mlx-serve, covering benchmarking and comparison traps (these cost real days).
Skill Claude CodeCodex
Timings measured 2026-07-16 on the M4 Max 128 GB, AFTER the stopallengines port-wait fix (before it, everything below was 2.2× slower — see the gotcha in Benchmarking).
Skill Claude CodeCodex
Use when the user asks to transcribe audio or video, inspect or manage MacParakeet meeting recordings, search prior dictations/transcripts, or give an AI agent local speech-to-text tools on Apple Silicon.
Skill Claude CodeCodex
Instructions for the OS-native taOS agent that operates the OS on the user's behalf. Covers the hard rule (drive the desktop only via POST /api/desktop/command + screenshot), opening and arranging apps and windows, projects, files, memory and notes, chat conventions, image generation, and answering the user. Load when…
Skill Claude CodeCodex
Procedures and architecture for contributing to taOS. The non-negotiable rules live in soul.md; this skill is the how.
Skill Claude CodeCodex
Use this skill when porting a NEW model to Core AI for the zoo — "convert to .aimodel", "add to the zoo", "why does my exported bundle produce garbage", "how do I gate a port", "what belongs in a zoo PR". Covers the oracle-first method, the two gates, compression choices, the device tier, and what a finished port must…
Skill Claude CodeCodex
Use this skill to rebuild, verify, or run a model from the Core AI model zoo (coreai-model-zoo) — any request like "convert Qwen3.5 for Core AI", "reproduce the gemma-4-E2B bundle", "which zoo model should I use for on-device chat / OCR / TTS / embeddings", "check whether this .aimodel was published correctly", or…
Skill Claude CodeCodex
Use this skill to generate well-branded interfaces and assets for Vocello (a local-first, Apple-native macOS voice-generation app — formerly QwenVoice). Contains essential design guidelines, colors, type, fonts, assets, and a Mac-app UI kit for prototyping voice-studio interfaces, marketing surfaces, and slides.
Skill Claude CodeCodex
This skill should be used when users want to fine-tune language models or perform reinforcement learning (SFT, DPO, GRPO, ORPO, KTO, SimPO) using the highly optimized Unsloth library. Covers environment setup, LoRA patching, VRAM optimization, vision/multimodal fine-tuning, TTS, embedding training, and…