MCP servers and skills: search, filter by where they run, and compare.
AI-driven climate forensics and real-time monitoring of El Niño. Leveraging Claude Managed Agents and 3D geospatial data to transform complex meteorological insights into actionable impact. Built for
SkillMIT★ 2Few starsNo key
The Universal AI Video Engine: Autonomous 3D Motion Graphics, VFX/CGI, Higgsfield Cinema DoP & Remotion NLE for Claude Opus, GPT, Gemini & AI Agents. Turn plain text into Hollywood-grade cinema via MC
SkillApache-2.0★ 1Few starsNo key
Own your LLM's web search: a local search->fetch->rank pipeline that replaces hosted web-search tools. Measured: matches hosted accuracy at 66% lower cost and up to 88% fewer tokens, plus a precision-
PythonMIT★ 60No key
Reliability eval harness for tool-calling LLM agents. Grounding, refusal, hallucination and trajectory metrics gated in CI, served over MCP, compared live across Claude models on AWS Bedrock.
SkillMIT★ 4Few starsNo key
Switchyard: a small local daemon that keeps one live session per coding agent (agy, Claude Code, OpenCode, Codex). Spawn, watch, steer, interrupt and stop them from one CLI, MCP bridge or dashboard, w
SkillMIT★ 2Few starsNo key
SkillBSD-3-Clause★ 4Few starsNo key
Unified research-data acquisition MCP — search & fetch datasets across Zenodo, DataCite, NCBI omics (GEO/SRA/BioProject), and literature (PubMed/OpenAIRE) behind one normalized model.
PythonMIT★ 4Few starsNo key
End-to-end AI pipeline that turns a topic into a YouTube-ready script, narration, and ~150 style-consistent images. Claude + Gemini + ElevenLabs, with a Flask UI and Telegram bot.
SkillMIT★ 1Few starsNo key
Single-user dashboard for net worth, Gmail, calendar and smart home. Go, React, SQLite, passkeys.
GoMIT★ 1Few starsNo key
Qwen3.8-Max rates within 30 Arena points of Claude Opus 5.5 for 3.3× less per output token..
Hosted by authorNo licenseFew starsNo key
claude-opus-4-6-high leads Arena coding at 1535. Anthropic has held #1 for 280 days, but Z.ai.
Hosted by authorNo licenseFew starsNo key
claude-opus-4-5-20251101-high-32k is #49 overall but #14 in coding. gpt-5.6-terra-xhigh is #52.
Hosted by authorNo licenseFew starsNo key
4-line Catppuccin Mocha statusline for Claude Code — shows dir, git, model, live spend, MCP server health, and real 5h/7d/Sonnet usage from Anthropic's OAuth API (same data as /usage)
SkillMITFew starsNo key
MCP server for Marcus Rug Intel — 19 tools for crypto token risk analysis, rug pull detection, and AI forensics. Works with Claude Desktop, Cursor, Windsurf.
PythonMIT★ 1Few starsNo key
Agent-driven geospatial analysis: Claude Opus 5 orchestrating QGIS through the Model Context Protocol to audit and correct a multi-criteria flood model.
MCPLink onlyFew stars
Real-time architectural sensor that helps AI agents close the feedback loop, enabling recursive self-improvement of code quality. Pure Rust.
SkillMIT★ 3.3kNo key
学术论文写作 agent 工具链：9 个 paper-* agent + 4 个自研 skill + paper-tools-mcp（24 个学术写作辅助工具），覆盖需求分析→文献→研究→撰写→审查→排版端到端流程。面向 Claude Code。License: Apache-2.0。
SkillApache-2.0★ 1Few starsNo key
Get the best of both worlds. Token-efficient autonomous coding pipeline for Claude Code. Codex plans, Opus reviews, Sonnet swarm implements, optional metric optimization (inspired by karpathy/autorese
SkillMIT★ 1Few starsNo key
Empirical benchmark comparing agent architectures (single-agent, multi-agent, adaptive) on ProgramDev-v0 and CyberGym tasks. Key finding: adaptive architecture > single-agent > fixed-pipeline multi-ag
SkillMIT★ 1Few starsNo key
See everything Claude Code knows — projects, MCP servers, sessions, costs — in one dashboard
Node.jsMIT★ 7No key
Local web dashboard for Claude Code — monitor rate limits, manage agents/skills/hooks/MCP servers, browse session history with conversation viewer, track daily activity and usage rankings. Built with
SkillMIT★ 1Few starsNo key
A local-first MCP server that answers, from real data on your machine: How many tokens have I used, from which client, model and session — and what did it cost?
Node.jsMIT★ 1Few starsNo key
Local, single-file HTML dashboard for Claude Code usage and estimated USD cost — token totals across all classes (input/output/cache), per-model spend at list API prices, live current-day data, MCP ac
SkillMITFew starsNo key
Who stands behind it
