Search

ADHD — a skill for coding agents. Fans out many divergent thoughts in parallel under different cognitive frames, scores them, prunes traps, and deepens the survivors. Tree-of-thought with pruning, built on the Claude Agent SDK. The no-brainer skill for cr

Agent skills published from my_agent_skill

Create a new moon-wave agent project

Hub Hosted Cockpit のローカル常駐 agent。Hub と outbound WSS で接続し、ローカルの tmux/pty を中継する。

See how much your AI coding agents actually cost — zero config, zero code changes.

Pi extension that adds /review via an isolated review subagent.

Generic Slack-chat plugin — connect one or more Paperclip agents to Slack via @mention or DM

The reliability layer for LLM tool calls. Intercepts every call, validates arguments against your schema, auto-repairs hallucinated args before execution.

Project-scoped GraphRAG memory for agents.

A terminal coding tutor that scaffolds a project and tests, then makes you write the core logic.

AWS CDK construct for deploying production-ready Bedrock AI agents with Lambda, DynamoDB, and OpenSearch

Differentiable cognitive primitives for agentic AI

Multi-axis bisect engine for finding LLM-agent regressions across prompt, model, tool-schema, and RAG-corpus changes

The golden standard of multi-agent orchestration — typed state graphs, pre-built agents, GroupChat, governed workflows, eval framework, SwarmTRM neural consensus engine, and the only framework built for regulated industries

Grid Memory — shared persistent memory for multi-agent teams

Transparent curl wrapper that records direct LLM API calls to Mesedi

Open-source AI agent operating system: one file-native brain (190+ skills, 180+ specialist agents) run across many sealed client arms, with per-client token attribution and opt-in budget caps. Organic, portable, MIT.

Shovs LLM OS is a local-first, open-source control center for autonomous agents.

Static linter for AI agent configs, tool descriptions, and system prompts with zero-LLM CI gating

The Memory Layer for Enterprise AI — persistent memory, observability, and 40% token cost reduction for any LLM stack