21Trust score 21 of 100, Limited
JSON Contracts
by json-contracts · TypeScript
JSON contract registry and validator for structured LLM and agent outputs.
Registry
Loading the directory
Category
AI and memory
MCP servers for embeddings, vector stores, model gateways, and long-term memory. MCPpedia lists 1744 published AI and memory MCP servers. The highest trust scores right now are Codebase Memory at 77, haiku.rag at 74, openInvest at 72. 751 listings speak a remote transport, which a hosted client can connect to with a URL. Each server page shows the license, the tools it declares, and a paste-ready config for Claude, Cursor, VS Code, and the other supported clients. Nobody can pay to move up the list.
21Trust score 21 of 100, Limited
Persistent memory for AI coding tools. 11 tools for any MCP-compatible AI tool.
Registry
21Trust score 21 of 100, Limited
53 tools for LLM Conveyors: job hunting, B2B sales, ATS scoring, resume tools.
Registry
21Trust score 21 of 100, Limited
Verifiable Memory for AI Agents — knowledge graph, FTS, Starknet proofs, local or cloud
Registry
21Trust score 21 of 100, Limited
Persistent codebase intelligence that gives AI assistants memory across sessions
Registry
21Trust score 21 of 100, Limited
Persistent memory for Claude Code — offline, single-file, hybrid search.
Registry
21Trust score 21 of 100, Limited
Persistent memory for AI coding agents. Hybrid search, auto-session capture, context routing.
Registry
20Trust score 20 of 100, Limited
Agentic Observability MCP
by mdfifty50-boop · JavaScript
Agent tracing, cost tracking, anomaly detection for LLM agents
Registry
20Trust score 20 of 100, Limited
MCP Server that renders JavaScript SPA pages and extracts LLM-ready Markdown content
Registry
19Trust score 19 of 100, Limited
Afterpaths
by burnssa · Python
Session memory for AI coding agents. Search past sessions, extract rules, and track what works.
Registry
18Trust score 18 of 100, Limited
knownissue
by gong8 · TypeScript
Shared debugging memory for AI coding agents
RegistryRemote
17Trust score 17 of 100, Limited
Recite MCP
by rivradev · Python
Local MCP server for Recite receipt processing with ledger and memory tools.
Registry
17Trust score 17 of 100, Limited
A paid remote MCP for OpenAI Codex agent coordination MCP, built to return verdicts, receipts, usage
RegistryRemote
17Trust score 17 of 100, Limited
jcodeharness-mcp
by clauxel · JavaScript
A paid remote MCP for OpenAI Codex harness MCP, built to return verdicts, receipts, usage logs, and
RegistryRemote
17Trust score 17 of 100, Limited
api-changelog
by weiseer · JavaScript
SDK breaking-change tracker for AI agents — Anthropic/OpenAI/LangChain/MCP/Mistral.
Registry
17Trust score 17 of 100, Limited
status-aggregator
by weiseer · JavaScript
Cross-vendor SaaS status aggregator for AI agents (Anthropic/OpenAI/AWS/GH/Stripe/CF).
Registry
17Trust score 17 of 100, Limited
A paid remote MCP for OpenAI Codex memory MCP, built to return verdicts, receipts, usage logs, and a
RegistryRemote
17Trust score 17 of 100, Limited
A personal RAG database you build from chat, so AI creates work that sounds like you.
RegistryRemote
17Trust score 17 of 100, Limited
A paid remote MCP for OpenAI Codex context compressor, built to return verdicts, receipts, usage log
RegistryRemote
17Trust score 17 of 100, Limited
llm-oracle
by weiseer · Python
Up-to-date LLM pricing + availability catalog with cited sources for AI agent routing.
Registry
17Trust score 17 of 100, Limited
casemgr-mcp
by mstang · JavaScript
Shared workspace your AI agent writes to. 184 tools, persistent memory, semantic search.
Registry
16Trust score 16 of 100, Limited
memoryoracle
by tooloracle · Dockerfile
MemoryOracle - 10 agent memory tools: vector store, recall, summarization, redaction.
RegistryRemote
15Trust score 15 of 100, Limited
Dynamic RAG Engine preventing AI hallucinations in Korean Finance and Crypto markets.
Registry
14Trust score 14 of 100, Limited
archetypal-ai
by archetypal-ai · Shell
Persistent memory for AI agents. recall, remember, checkpoint — soul preservation.
Registry