Amal-David

Publish local HTML, Markdown, or built static web projects with Pagecast as shareable public URLs. Use whenever Codex creates or finishes an .html, .htm, .md, .markdown, or static build output that a person could share (a report, plan, doc, dashboard, or analysis) — proactively offer to publish it without being asked — and whenever the user asks to publish, share, make a public link for, or send a local report/doc/dashboard/web project from terminal, Codex CLI, or Codex desktop.

Help publish local HTML, Markdown, or built static web projects by asking whether to use Claude Code Artifacts or Pagecast/Cloudflare Pages when the user's publish request is ambiguous. Use when a report, plan, doc, dashboard, analysis, or static build output should be shared from Claude Code, especially when the user says publish, share, make a public link, or send this. If the user explicitly says Pagecast, publish with Pagecast.

Switch from Claude Code to Codex without losing context. Packages the exact active session, recent decisions, and git state, then opens Codex. Use when Claude hits usage or context limits, changes models, or the user invokes /handoff or asks to continue in Codex.

Support the standalone /handoff command that resumes Claude Code sessions in Codex with Fable/Opus model lineage, limit/fallback detection, policy-neutral context, a configured target model, private context files, and git state. Use when maintaining, explaining, or debugging the command installed under ~/.claude/commands/handoff.md.

Hand off a live Claude Code conversation to Codex by packaging the local Claude transcript, writing a compact digest, and launching or preparing an interactive Codex terminal session. Use when the user asks to continue in Codex, is hitting Claude limits, or wants a Claude to Codex handoff.

Support files for the standalone /handoff command. Use when maintaining, explaining, or debugging the Claude to Codex handoff command installed under ~/.claude/commands/handoff.md.

Guides and validates architecture-aware ports of PyTorch/Hugging Face models to Apple MLX, inspects existing local MLX projects, and plans evidence-gated optimizations for Apple Silicon. Use when the user asks to run, port, convert, inspect, quantize, benchmark, or fix a model (LLM, VLM, audio/TTS/ASR, diffusion, SSM, MoE) for MLX, MLX-LM, MLX-VLM, MLX-Audio, or a Mac - e.g. "port this HF model to my Mac", "inspect this MLX app", "run Qwen on Apple Silicon", "convert these safetensors to MLX", "make this faster on my M3", "fix NaN in my MLX port", "speed up prefill / KV cache / speculative decoding", "publish an MLX checkpoint". Also use mid-task when a config.json, safetensors index, weight-shape or tokenizer mismatch, or Metal kernel question appears. Do not use for CUDA-only optimization, non-Apple hardware targets, or general PyTorch/ML questions with no MLX or Apple Silicon connection.