
from everything-openai-codex73
Per-project persistent memory and session tools for Codex: register projects, save session summaries, resume context, and manage project contexts via determinis
Context Keeper (ck) provides persistent, per-project memory and deterministic session tooling for OpenAI Codex workflows. It organizes project context under ~/.codex/ck, stores a structured context.json as the source of truth, and exposes a set of Node.js command scripts (init, save, resume, info, list, forget, migrate) that the agent runs verbatim to register projects, capture session summaries, resume full briefings, and migrate older formats. The skill enforces a safe workflow: scripts auto-confirm actions and the agent displays script stdout directly to the user for transparency.
Use ck when you want the agent to load and persist project-specific context across sessions, capture concise session summaries and next steps, or present a full briefing for a named project. It's appropriate for development workflows where reproducible, script-driven saves and resumes are preferred over ad-hoc memory edits.
Best fit for Codex-style and CLI-integrated assistant harnesses (OpenAI Codex, Codex CLI wrappers, Cursor, Gemini CLI), and any agent setup that can run local Node.js scripts and display stdout.
This skill has not been reviewed by our automated audit pipeline yet.
Motion Foundations
Foundation motion tokens, spring presets, accessibility rules, and SSR-safe animation patterns for React/Next.js using motion/react.
VideoDB
Tools for ingesting, indexing, searching, editing, and transcoding video/audio (files, streams, desktop capture) with timeline editing, transcripts, and alerts