Freeze Claude Code's prompt prefix so DeepSeek's automatic cache always hits — alignment proxy + coalescing + keepalive, installable as a CC plugin. Measured…
Healthiest
// sorted by health — what's thriving right now
// observatory (5535)
给 AI agent(Claude Code / openclaw / Codex)用的 VPS SSH 管理技能:统一密钥、别名管理、一键下发公钥、安全关闭密码登录、批量探活
Claude Code skill that turns any topic, conversation, or file into a single-file, beautifully typeset HTML document - trigger with /paper
洁癖 Skill: audit Codex, Claude Code, and Cursor skill usage
One natural-language policy file. Prompt-based enforcement hooks for Claude Code, Codex, and Cursor.
Doubt, until what remains is certain. An MCP server that recursively doubts every plan decision — grounding each in real evidence or asking you — until it co…
CLI tool wrapping claude code to help understand cost breakdown
AI-native methodology for Go backend microservices: rebuild your codebase from docs/ + .claude/skills/ alone — correct, IO-efficient, architecturally sound b…
Claude Code skill for systematically reducing VirusTotal/EDR detection rates on compiled Go binaries via PE structural analysis and disciplined A/B testing.
Run Claude Code subagent/UltraCode fan-out on DeepSeek v4-flash via the Reasonix CLI — keep Claude as the main agent, route fan-out lanes to a local reasonix…
Turn an agent's work: progress, reviews, research, comparisons into a single self-contained, reviewable HTML page (a work canvas). An Agent skill.
A physical traffic-light status indicator for Claude Code, Cursor, and AI coding agents
Multi-model consensus planner for coding agents — Claude + Codex + DeepSeek draft a plan independently, cross-verify, and must agree before any plan is emitted.
One-click local backup for Codex and Claude Code memories
Claude Code skill: turn one idea into a finished vertical cartoon episode + social carousel (Higgsfield + ElevenLabs + ffmpeg)
Simple prompts and habits to catch an AI lying, guessing, and telling you what you want to hear - and make it much harder to get away with. Plus Claude Code…
Engineering Loop — a CLAUDE.md workflow template that turns AI coding agents into structured engineers: 9-phase process from task intake to goal verification…
Open-source Agent Operating System
One CLAUDE.md file. Keeps Claude responses terse. Reduces output verbosity on heavy workflows. Drop-in, no code changes.
Use any LLMs (Large Language Models) for Deep Research. Support SSE API and MCP server.
Visual Skills Pack for Obsidian: generate Canvas, Excalidraw, and Mermaid diagrams from text with Claude Code
Open Source ClaudeCode Leaked. Same functionalities as Claude Code by Anthropic. Written in Rust, Ready to use immediately for free.
The local-first LLM Wiki: open-source knowledge graph builder, RAG knowledge base, and agent memory store. Built on Andrej Karpathy's pattern. An Obsidian al…
claude-code-api — a Claude ecosystem project on GitHub.
Opus에 근접한 품질을 Sonnet 단가에 — 가성비 AI 운영 모델을 직접 구축한 Claude Code 프로젝트
A Claude Code hook that blocks your prompt until you do your push-ups, counted live via webcam. Random reps, session-persistent debt (no closing the tab to s…
Loop until it's better — drop-in agentic loops (autoresearch, scientific writing, data analysis, code/SQL/prompt optimization, red-teaming) as open-standard…