AI & ML MCP Servers
LLM providers, agents, models, embeddings, inference, and machine learning workflows.
Abe
Published by GitHub account flowinfosystems-index
Abe is a free local guardrail that checks an AI agent's proposed action against your YAML rules and returns ACT, BLOCK or ESCALATE before it runs.
ContexteTech
Published by GitHub account contexte-tech
ContexteTech lets an assistant search and load AI datasets, prompts and few-shot contexts from its catalogue, fill prompt templates, and estimate model running.
Where My Hug
Published by GitHub account bikemeardsley
Searches Hugging Face, Kaggle, Zenodo, ModelScope and Roboflow Universe for models and datasets matching each part of an ML job, then verifies every pick live.
CanaryGuard 402
Published by GitHub account ak10082247-max
Paid remote MCP server that sanitizes untrusted text for prompt injection and audits Base wallets, tokens, contracts and MCP tool definitions.
Defigent
Published by GitHub account Firrton
A remote MCP endpoint giving an assistant an agent messaging inbox with signed messages, an ERC-8004 directory, and status and text hashing for Monad.
Codeman
Published by GitHub account Ark0N
Drives Codeman's AI coding agent sessions (Claude Code, Codex, Gemini CLI and others) in tmux: start, prompt, wait, read output, interrupt and delete.
PromptBase
com.promptbaseOfficial publisher
PromptBase is a remote MCP server that lets an assistant search and use a large library of AI prompts and skills for images, video, writing, games, apps and web
Agenlify AI Agent Directory
ai.agenlifyOfficial publisher
Searches the public Agenlify AI agent directory by task, product or domain, returning pricing, setup, limitations and official sources.
Poof
Published by GitHub account poof-bg
Poof's MCP server lets an assistant remove backgrounds from images, videos and animated GIFs through the Poof API, returning cutouts, mattes or stickers.
Parrely
Published by GitHub account bliongosari1
Parrely is a remote MCP server that gives AI agents shared memory, live work status, agent profiles and messaging so multiple agents can work together.
Canli Quant
Published by GitHub account arhancanli
Local MCP server exposing 236 quant finance calculators — performance, options, indicators, econometrics, portfolio, sleeves and labeling — on data you supply.
SpanHound
Published by GitHub account abhishekash
SpanHound captures local AI agent traces and exposes read-only MCP tools to inspect failures, approvals, token usage and run comparisons.
Threat Rules
Published by GitHub account GuardBee
Local MCP bridge to Agent Threat Rules (ATR) that lets an assistant score text, files and agent events for prompt injection, tool poisoning and MCP attack…
Model Scanner
Published by GitHub account GuardBee
Scans ML model files and Hugging Face repos for supply-chain risks such as dangerous pickle globals, without loading or executing them.
Memory Poisoning Scanner
Published by GitHub account GuardBee
Scans AI agent source code for memory-poisoning patterns, where untrusted input is written into persistent memory or recalled memory is fed back as trusted…
Llm Redteam
Published by GitHub account GuardBee
Runs canary-based jailbreak, prompt-extraction and obfuscation probes against a live LLM endpoint or chatbot you control and reports which guardrails were…
Folkmoot
com.salukilabsOfficial publisher
Lets AI agents debate proposals, vote with sealed ranked ballots, and get second opinions from other models in public halls.
Bike4Mind — the open-core AI workbench
com.bike4mindOfficial publisher
A read-only remote MCP server that gives an assistant Bike4Mind's model catalog, product list, comparisons, licensing facts, blog posts and product updates.
Prompt Guard Mcp
Published by GitHub account robertoyanguas-ops
A remote MCP server that scans untrusted text such as web pages, emails or tool outputs for hidden prompt injections and returns sanitized text.
Json Repair Mcp
Published by GitHub account robertoyanguas-ops
Remote MCP server that repairs truncated or malformed JSON from LLM output and reports an agent's prepaid wallet balance and usage.
Archestack
Published by GitHub account archestack
Read-only access to Archestack's rankings of open-source AI projects: search, compare, check deploy needs and find alternatives to paid tools.
Teardrop
Published by GitHub account HASAVAGE
Teardrop is a paid MCP gateway that gives agents metered platform tools, agent-to-agent delegation, and scheduled or event-triggered agent runs.
PurRAssist
Published by GitHub account Blackcat-Informatics
A local stdio MCP server that repairs imperfect AI output — malformed JSON, tool calls, names, paths and Markdown edits — and returns validated results with…
P0 Agent Economy Index
dev.workers.x402-bazaar-rank-worker.x402-bazaar-rankOfficial publisher
Open index of the agent economy: search paid x402 services and buy per-call data such as SEC filings, wallet risk, OFAC and EVM reads.