MCP servers and skills: search, filter by where they run, and compare.
1–18 of 18 tools for “test-time-scaling”
Clear all🚀 MassGen is an open-source multi-agent scaling system that runs in your terminal, autonomously orchestrating frontier models and agents to collaborate, reason, and produce high-quality results. | Jo
MCPLink only★ 1.1k
Atom of Thoughts (AoT) MCP is a server that decomposes complex problems into independent atomic units of thought, using the dependencies between these units to deliver more robust reasoning and valida
Node.jsMIT★ 61No key
A Claude Code skill that fans one hard question out to several frontier models (GPT-5.5, Gemini, Kimi, GLM) blind and in parallel, then has Opus 4.8 judge and synthesize one grounded answer. Multi-mod
SkillLink onlyFew stars
MCP server for searching 600,000+ CS/AI research papers with citation graphs, full text, embeddings, and BibTeX. Works with Claude Code, Cursor, and any MCP client. No signup required.
Node.jsMIT★ 13No key
Sovereign Frontier AI Lab architecture. System 2 RLVR reasoning, DeepSeek-R1 GRPO self-training, deterministic AST sandboxes, and native Windows OS computer-use ($0 capital, zero third-party APIs).
SkillApache-2.0★ 1Few starsNo key
Reduce coding-agent delivery cost by proving one complete path before scaling implementation.
SkillMIT★ 2Few starsNo key
Medium API in Python & JS: scrape Medium articles by tag, author, publication or search, with claps, reading time and full text as Markdown for RAG. n8n + MCP.
SkillMITFew starsNo key
GenPark AI Agent Skill - DeepSeek-R1 reasoning verification, test-time compute scaling, multi-head latent attention (MLA), and MoE dispatch.
SkillLink only★ 7
[ACL 2026 Demo] ScaleBox: Enabling High-Fidelity and Scalable Code Verification for Large Language Models
SkillApache-2.0★ 8No key
An MCP (Model Context Protocol) server that generates advanced mock 2D PNG assets for games prototypes — directly from any MCP-compatible AI client.
Node.jsMIT★ 6No key
SOTA deep research plugin for Claude Code — multi-agent, multi-perspective, citation-verified research engine
SkillMIT★ 1Few starsNo key
PythonMIT★ 1Few starsNo key
Claude Code skills for paid search on a budget too small for the standard playbook. Budget floors, consolidation, and refusing to decide on thin data.
SkillMITFew starsNo key
High-Trust Open-Core framework for Windows AI Agents. Crushes latency to 3.9ms, enables background UIA invocation, and protects enterprise RPA workflows.
SkillMIT★ 3Few starsNo key
Train and evaluate syndrome-based neural decoders on your favorite linear code
SkillMIT★ 2Few starsNo key
SkillMIT★ 1Few starsNo key
Delegation and completion discipline for autonomous agents: one Claude Code skill, two halves, one ledger
SkillMITFew starsNo key
Who stands behind it
