Odel

AI & agents

Memory, payments and tools for AI agents, other agents to work with, and gateways to more tools.

12,705 servers6,739 publishers11 tags

All of AI & agents

By GitHub stars, at most two per publisherRemoteLocal
MCP server: One MCP server.
Robot Resources Router
@robot-resources · External

Intelligent LLM routing proxy — 60-90% cost savings by auto-selecting the cheapest model

7% chance this is malicious. Assessed by AI, which can make mistakes.Can move money, run code, delete data or send messages for you. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Deprecated by the publisher: Replaced by in-process router (@robot-resources/router). A new MCP server tied to the public /v1/route endpoint landed in PR 8 (multi-agent compat pack, #191). Install: npx @robot-resources/routerFree to use.
MCP server: One MCP server.
state-trace
@razroo · External

Bounded working memory for coding agents — typed nodes, current-vs-stale state, MCP-mountable.

4% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.Free to use.
MCP server: One MCP server.
Agent Toolbelt
@rccola990-cloud · External

MCP tools for AI agents: fetch any URL as clean text, live crypto prices, review & bid drafting.

8% chance this is malicious. Assessed by AI, which can make mistakes.Only reads or looks up information. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
Token Optimizer
@rccola990-cloud · External

Cut LLM token costs: count tokens, estimate cost, slim prompts, and pick the cheapest capable model.

4% chance this is malicious. Assessed by AI, which can make mistakes.Only reads or looks up information. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
AgentAuth
@raditotev · External

Identity and auth for AI agents: credentials, tokens, delegations, and permission checks.

6% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.Free to use.
MCP server: One MCP server.
modelmux
@pure-maple · External

Model multiplexer — route tasks to Codex, Gemini, and Claude CLIs with smart routing.

5% chance this is malicious. Assessed by AI, which can make mistakes.Can move money, run code, delete data or send messages for you. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.Free to use.
MCP server: One MCP server.
Pretorin ComplianceRun by its publisher: served from their own domain, namespace checked by the registry.
@pretorin-ai · External

Access Pretorin controls, evidence, narratives, and local Plan-backed Campaigns from AI tools.

11% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.Free to use.
MCP server: One MCP server.
MemoraEU
@pquattro · External

Zero-knowledge persistent memory layer for AI agents. Encrypted client-side, hosted in France.

6% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.Free to use.
MCP server: One MCP server.
Agentos
@pdaxt · External

AI agent fleet orchestration runtime. Spawn, coordinate, monitor autonomous agents.

6% chance this is malicious. Assessed by AI, which can make mistakes.Can move money, run code, delete data or send messages for you. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.Free to use.
MCP server: One MCP server.
Dark Search Protocol
@parnasheps-eng · External

161 dark data sources, 25+ domains, 6-step pipeline, Streamable HTTP + stdio

11% chance this is malicious. Assessed by AI, which can make mistakes.Only reads or looks up information. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
NIRA REPL MCP
@pachote · External

Secure code execution sandbox for Claude — run Python, JavaScript, and shell commands

10% chance this is malicious. Assessed by AI, which can make mistakes.Can move money, run code, delete data or send messages for you. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.Free to use.
MCP server: One MCP server.
NIRA Video Production MCPRun by its publisher: served from their own domain, namespace checked by the registry.
@pachote · External

Video production MCP: 55 tools — edit, grade, TTS captions, beat-sync, master, AI generate.

5% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.Free to use.
MCP server: One MCP server.
FasciaRun by its publisher: served from their own domain, namespace checked by the registry.
@orcar-studio · External

Expose Fascia-deployed Tools and Entity schemas to AI agents via MCP

4% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
ClawTube
@oogdevs · External

Minimal MCP server for ClawTube — a small, auditable control plane for skills and agents.

5% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.Free to use.
MCP server: One MCP server.
Opencontext Team MCP
@opencontext-team · External

An MCP server that provides visual memory and context storage with knowledge graph capabilities

9% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
Novyx MCPRun by its publisher: served from their own domain, namespace checked by the registry.
@novyxlabs · External

Persistent memory for AI agents with rollback, audit trails, and semantic search.

5% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.Free to use.
MCP server: One MCP server.
Product Lock
@nickjazz · External

Product Lock tools for AI agents: generate, review, validate, and score locks.

5% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
Oroute
@netclaus · External

Auto-route AI requests to 13 models (Claude, GPT, Gemini, Qwen, DeepSeek). 57% cost savings.

8% chance this is malicious. Assessed by AI, which can make mistakes.Can move money, run code, delete data or send messages for you. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
Memento Protocol
@myrakrusemark · External

Persistent memory for AI agents. Store what matters, recall by meaning, skip the rest.

5% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
Splitifi Intelligence MCP
@mysplitifi · External

Outcome predictions and judge profiles from 2,741 ML models trained on 30.3M court records.

6% chance this is malicious. Assessed by AI, which can make mistakes.Only reads or looks up information. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
MostAI
@most-ai-com · External

Generate images, video, music, speech and text with 60+ models through your MostAI account.

7% chance this is malicious. Assessed by AI, which can make mistakes.Creates or changes data. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.Free to use.
MCP server: One MCP server.
AgentVault
@motiveflowllc · External

Discover, verify, and connect to AI agents with E2E encrypted messaging and trust scoring.

7% chance this is malicious. Assessed by AI, which can make mistakes.Can move money, run code, delete data or send messages for you. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
Mova Flat RunnerRun by its publisher: served from their own domain, namespace checked by the registry.
@mova-compact · External

Governed AI workflows with human approval gates and audit trails for Claude and MCP clients.

6% chance this is malicious. Assessed by AI, which can make mistakes.Can move money, run code, delete data or send messages for you. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.Free to use.
MCP server: One MCP server.
MOVA MCP Connector
@mova-compact · External

Thin MOVA MCP proxy for bindings, transport resolution, and upstream tools/call execution.

5% chance this is malicious. Assessed by AI, which can make mistakes.Can move money, run code, delete data or send messages for you. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.Free to use.

More topics

Sales & customersMarketing & advertisingShopping & e-commerceBusiness financePersonal finance & investingLegal & governmentSecurity & privacySoftware developmentCloud & IT operationsData & webWork & productivityCommunicationDocuments & writingDesign & mediaNews & informationScience & researchHealth & fitnessEducation & careersTravel & placesHome & propertyEntertainment & hobbiesIndustry & transport