Agent skills for ChatGPT

Read and act on the user's Gmail, Google Calendar, and Google Tasks through per-user OAuth.

Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists…

Full safety mode: destructive command warnings + directory-scoped edits. (gstack)

Execute Hugging Face Hub operations using the `hf` CLI.

Fast BPE/WordPiece tokenization and custom vocab training.

Use the HyperFrames CLI development loop: init, add, catalog, capture, lint, check, snapshot, compare, grade-compare, preview, play…

Fetches and references LangGraph Python documentation to build stateful agents, create multi-agent workflows, and implement…

Recent community and social trend research over the last 30 days.

Build leveraged buyout workbooks with IRR/MOIC in Excel.

lm-eval-harness: benchmark LLMs (MMLU, GSM8K, etc.).

Open Design's unit-economics and BYOK cost model: the assumptions, the sensitivity, and why it scales.

MemPalace — Local AI memory with 96.6% recall.

Serverless GPU cloud for ML jobs and model APIs.

Next.js 16+ and Turbopack — incremental bundling, FS caching, dev speed, and when to use Turbopack vs webpack.

npm registry ops: login, whoami, names, publish; 1Password tmux.

Browse and run any of the ~350 models served by the local cli-proxy-api (all OpenRouter models plus Codex).

Read, extract (text/tables), create, merge/split/rotate, watermark, encrypt,

Screenshot and visually read any web page or document using pixelshot.

Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction…

Persistent browser and Electron interaction through `js_repl` for fast iterative UI debugging.

Build a throwaway prototype to answer a design question.