Coding Assistants

A Claude Code / Codex skill that automates the entire film pipeline, using Remotion to turn research and narration into deterministic 720p motion graphics with TTS voiceover.

It reads the session logs already sitting on disk and turns them into a cost breakdown by model, day, and project—no API keys, no network calls.

It stores AI agent memory as plain Markdown inside your repository, replacing expensive vector databases with sub-300-microsecond local search and cutting token bloat by 80%.

A configuration pack that assigns cheaper Luna subagents to Codex grunt work while reserving GPT-6 Astra for orchestration and review, so you pay for capability only where you need it.

It borrows the sign-ins you already have to turn scattered AI usage limits into a persistent screen-edge dashboard.

unlazy forces AI agents to prove their work through runnable acceptance gates instead of trusting their own exit reports.

A local-first system that treats your authorized WeChat database as a read-only intelligence source, generating daily briefings and opportunity tracking through natural-language Codex skills.

Most browser agents chat at you; this one returns editable spreadsheets, slides, and sites you can actually use.

Because writing LinkedIn content that sounds human is tedious, this repo packages 11 prompt-engineered skills that let your AI agent draft, audit, and humanize posts—then wait for your approval before publishing.

SoL-Pi packages four auto-researched mechanisms that cut token traffic and repeated turns in the Pi agent without hiding evidence or skipping verification.

It stops long-running agents from drifting by forcing them to verify progress through independent roles before every new round.

It’s the public lobby for a proprietary AI workbench, collecting vision docs and ecosystem partnerships while keeping the runtime behind closed doors.

agent-memory persists what an agent learns across sessions as plain Markdown files with local ranked retrieval, so Claude Code, Codex, and any shell-capable host can share one store without API keys.

A reference architecture for Claude-powered shopping and merchant agents that keeps every write staged and every brand fictional.

Moli is a standalone Rust headless browser that avoids continuous rendering by computing layout and paint only when an operation explicitly demands geometry or pixels.

This plugin watches your Claude Code sessions, turns recurring patterns into native skills, and quietly archives the ones you stop using.

Because LLMs confidently invent struct offsets and behavior from binaries, this toolkit makes them prove every claim against the actual bytes.

Open-source Claude skills that run SEO and generative-engine optimization on your actual Google data instead of scraped estimates.

It’s a from-scratch Rust replacement for Docker Desktop on Mac, with a side of Firecracker microVMs for your AI agents.

A read-only MCP bridge that turns your paid ChatGPT web session into the planning and review brain for Codex, while execution stays locked to the agent.