Browse agent skills

MindExec Direct Remote agent

AI logo animation skill: turn raster logos into smooth SVG animation, animated HTML demos, GIF/video previews, and motion QA evidence.

Structure Gateway HTTP Client - A flexible HTTP client for communicating with Structure API Gateway

Agent skills for Wallive pet packages, Island shader packages, and plugin authoring workflows.

yunomi 🍵 — Human-in-the-loop approval gate for AI coding workflows. The AI serves its report; you review and approve. (formerly reviw)

MedSci Skills — a medical/scientific research skill suite for AI coding agents (Claude Code, Codex, Cursor, Copilot). The npm package is a terminal-friendly installer shortcut; the canonical distribution remains the GitHub repository and the Claude Code p

Visible, controllable multi-agent CLI workspace for Codex, Claude, Gemini, Kimi, MiMo, OpenCode, Antigravity, and Droid.

Drop-in SDK for the Canopy proxy. Routes OpenAI and Anthropic calls through Canopy for cost observability.

Local daemon and CLI for forwarding agent lifecycle events to Runlight.

Turn macro intuition into validated Dynare .mod files — a coding agent skill for DSGE modeling.

Local control plane for coordinating Claude Code, Codex, and other coding CLI sessions.

EvoKit — Self-Evolving System Framework for AI Coding Assistants

Claude Code skill: render Markdown into strictly-formatted Word docs (skeleton + profile modes)

Abkus Code — autonomous development agent. Like Claude Code but open-source.

Shared HTTP-client plumbing for the hive CLIs (login/cookie, JSON fetch, hidden prompt)

Official JavaScript SDK for TactiChat.

Pack any project folder into a compact, AI-readable XML snapshot — strip noise, keep signal, cut context by ~50%

CLI for the AIGentDocs docs-as-code standard: scaffold, validate, and maintain AI-agent-ready project documentation. Deterministic — no LLM involved.

A Python library for generating Final Cut Pro XML (FCPXML) timelines programmatically, designed to help video editors collaborate with LLMs on automatic edits.

An agent harness for TypeScript repositories: one command that is the definition of done, plus mechanically enforced module boundaries.

Simplified ML and NLP implementations — no dependencies except numpy

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

Zero-config Qdrant database Workbench — An AI-native GUI for the Qdrant vector database. Browse schemas, use the REST API console, leverage RAG tools, perform hybrid search, and import CSV data. Launch instantly with npx qdrant-db-workbench.

Companion CLI for the QUALITY.md file format and /quality agent skill, used to evaluate and improve AI assistant projects and harnesses.

Zero-config Pinecone DB Workbench — AI-native vector database GUI. Indexes browser, semantic search, metadata explorer, embeddings viewer & query console. Run with one `npx pinecone-db-workbench`

Visualize peer-review comments for a rebuttal: a self-contained, offline HTML page of the original reviews highlighted in place, colour-coded by action, grouped into priority-ordered rebuttal points each with an editable draft. Ships as a Claude Code skil

Pure FlatInk engine: model evaluation, expressions, timeline, geometry, layers. No canvas, no clipper. Shared by the player and the compiler.

The lightweight FlatInk player: load a .flatpack and play it in a <canvas>. No editor, no heavy deps.

Web search tools for agentkit-js — Tavily, Brave, Perplexity adapters

Browser automation tools for agentkit-js — Playwright + CDP-bridge for navigate/click/fill/screenshot/extract