Agents

It gives Claude Code and Codex a motion-design vocabulary—106 shot recipes, 161 previews, and a Remotion template—so they can direct cinematic product videos instead of generic slideshows.

This repo open-sources a full Chinese textbook on AI agent engineering—complete with Markdown source, compiled PDF, and runnable Python demos for every chapter.

It bundles the entire AI drama workflow—script, storyboard, voice, and final cut—into a single pipeline you can host yourself.

wigolo gives AI coding agents a local, keyless web research layer that runs entirely on your machine and bills nothing per query.

A skill that spawns parallel reasoning processes under distorted cognitive frames, then scores and prunes them with a separate critic pass.

An agent skill that forces you to approve the visual metaphor and static layout before it spends your Gemini credits on video generation.

LiveAgent is a local-first desktop client that gives LLMs real file-system and shell access, plus an optional gateway for remote browser control.

It bridges one-shot AI video generators and traditional timeline editors by letting conversational agents propose real, undoable edits to a local multitrack project.

Raven wraps agents in durable memory and self-refining skills so workflows survive past the chat session.

AOS Community Edition is a Rust-based agent operating system that lets agents inspect the runtime, spot missing capabilities, and forge their own least-privilege extensions.

iFixAi runs up to 32 inspections against any LLM or agent and returns a letter-grade scorecard in minutes, using a separate provider as judge so the model isn't grading its own homework.

It turns draw.io into a live artboard for Codex, letting an AI agent build and refine scientific figures one editable shape at a time.

NitroStack treats MCP servers as first-class backends, not afterthoughts held together with glue code.

Agent skills that turn your existing LLM coding assistant into a parallelized SAST scanner without extra tooling.

Flint compiles compact, human-editable chart specs into polished Vega-Lite, ECharts, or Chart.js visualizations so agents and developers can skip hand-tuning scales, axes, and spacing.

A local-first personal assistant that unpacks the four pillars of agent engineering—harness, loop, memory, and eval—into plain Python you can read in an afternoon.

Hyperresearch turns Claude Code into a tiered, multi-agent research pipeline that persists every source in a searchable, compounding vault.

pilotfish keeps Claude Fable 5 in the boardroom by delegating the grunt work to cheaper Sonnet and Haiku subagents.

This repository packages the public metadata that lets MCP clients discover and query Vybe’s live Solana API.

It turns Claude, Codex, or OpenCode into hyperparameter optimizers that actually read your parameter docs before proposing the next trial.