MCP Servers

Persistent memory and codebase context for AI coding agents. FAISS vector store, AST reverse index with O(1) symbol

Encrypted secrets vault with MCP for AI agents. Secrets resolved at runtime, never leaked to LLM conversations.

VMware vSphere compliance and hardening — read-only baseline scanning plus drift detection across CIS, vSphere SCG, China DJCP 2.0, and PCI-DSS frameworks. 6 read-only tools with LLM-powered remediation suggestions (apply-side gated through vmware-pilot approval workflow).

MCP stdio interceptor (@jamjet/mcp-shim) that applies one YAML policy file (block / requireapproval / audit / budget cap) to tools/call requests before they reach the real MCP server. The same policy also runs in Claude Code PreToolUse hooks (@jamjet/claude-code-hook), OpenAI Agents SDK guardrails (@jamjet/openai-guardrail), and JamJet's Python/TS SDKs — jamjet audit show tails every decision across every adapter from ~/.jamjet/audit/.

🎖️ - Official MCP server for Forge, Voxell's hosted text-embedding API. Generate vector embeddings (turbo 1024d, pro 2560d, ultra 4096d; Matryoshka truncation) for semantic search and RAG. npx -y @voxell/forge-mcp

Semantic search over 600k+ CS/AI papers with citation-graph traversal, full-text extraction, embeddings, and BibTeX export. Works anonymously or with a free key. Install: npx scholar-feed-mcp init.

Unified AI model API for 130+ models across 18 providers. Browse models, check pricing, create image/video/music/audio tasks, poll results, check balance, and call LLM endpoints. Free catalog tools work without an API key. npx @runapi.ai/mcp

Production-grade persistent memory service for AI agents. Semantic search with recency decay (configurable α blend), local ONNX embeddings (zero API cost), background memory summarisation, and Prometheus/Grafana observability. docker compose up — no cloud dependencies.

Lets the LLM reach you on your phone: live, interruptible voice calls and tiered alerts, using free self-hosted pieces (SIP via pjsua2/Linphone + whisper.cpp STT + macOS say TTS). No paid telephony, no extra API key for the conversation.

Service-boundary-aware document exchange center for coordinating heterogeneous LLM code agents. Versioned Markdown store, pub-sub notifications, diff-aware updates, and lifecycle stage tracking. Each service registers as a sub-project; agents subscribe to cross-service document changes and receive structured diffs via MCP.

Track LLM token costs across Claude, GPT and Gemini. MCP server + CLI with optimization hints and $ savings estimates.

Connect your agent to any HTTP API on the fly — discovers + maps any REST API once, then fetches typed data deterministically (no per-call LLM). Self-hosted MCP server (uvx --from 'liquid-api[mcp]' liquid-mcp); works with OpenAI/Gemini/Anthropic/local or any provider via LiteLLM. Open source (AGPL).

Browse 160+ LLM models (OpenAI, Anthropic, Google, Qwen, DeepSeek, …) with live pricing — no API key required for catalog tools. Routes chat completions through the OrcaRouter gateway with automatic fallback. npx -y @orcarouter/mcp.

Privacy-preserving AI gateway. Sanitises PII (German + English; Microsoft Presidio + custom recognisers) before prompts reach Anthropic / OpenAI / your LLM, then emits a signed cryptographic certificate per call (Ed25519 + RFC 3161 timestamp + Sigstore Rekor anchoring). EU GDPR + AI Act ready. Free tier 500 calls/month, BYOK. Install: npx -y @lucairn/mcp-server. Docs: https://lucairn.eu/developer/mcp.

Windows desktop automation for LLM agents with entity-based actions instead of coordinate-only clicking. Uses UIA, CDP, screenshots, keyboard/mouse/clipboard, and terminal control, plus entity leases, verified delivery, causal context, and interaction memory to reduce silent UI automation failures.

Synthetic monitoring for LLM inference endpoints. Measure TTFT, latency, throughput, and errors across OpenAI, Anthropic, Google, Azure, Bedrock, and local servers (vLLM, SGLang, Ollama). CLI + MCP server with Prometheus and OpenTelemetry export.

Deterministic knowledge graph MCP server. Single binary, zero LLM/embedding calls in the bridge, BLAKE3 state hashing, canonical KREX export for byte-identical audit. Local-first via redb (ACID). Alpha.

Personal knowledge graph for Claude on Neo4j. Typed entities (Person/Project/Decision/Preference/Concept), bi-temporal edge validity (validat/invalidat), per-type weighted decay, contradiction detection, and a nightly "dream" extraction process. Local bge-small-en embeddings, no API keys. Optiona

LLM-agent-native backtest framework for SHFE daily futures research. 23 MCP tools (strategy validation, scaffolding, indicator catalog, error-code lookup), 22 in-package skills, 32 catalogued error codes, working strategy templates. pip install echolon then claude mcp add -s user echolon -- echolon-mcp.

Deterministic batch tools so LLM agents stop next-token-guessing dates and math. Rich now() snapshot (18 fields), calendar(ops) batch dispatcher (diff/until/since/add/weekday/businessdays, natural-language parsing), calc(expressions) Python eval with math+stats pre-loaded, and Pint-based unit conversion. One wiring for da

🎖️ - MCP server for Jungle Grid, an agentic GPU execution layer that lets AI agents estimate, submit, monitor, and fetch logs for inference, training, fine-tuning, and batch workloads.

Agentic data quality framework that runs structured rules against warehouses (DuckDB, BigQuery, Athena, Databricks, Postgres), diagnoses failures with LLM root cause analysis, and proposes SQL remediations. Every LLM decision is audit-logged with cost and latency.

LLM character consistency engine — generates structured JSON constraints from 4 multiple-choice questions about an AI's psychology. Drop the JSON into any LLM's system prompt to prevent persona drift; reduces inference cost from retries. 160,000 personality patterns; works with any LLM.

Run DeepSeek as a real sub-agent inside Claude Code / Codex CLI. DeepSeek gets its own 7-tool agent loop (Read/Write/Edit/Bash/Glob/Grep/NotebookEdit) in a sandboxed workspace, not just a single LLM call. Includes token-based command sandbox, pre-flight web-search pattern, and cross-platform installer.