MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
🎓 ICIL — Intelligence Campus Interactive Library. 243 courses · 32 faculties · 10 MCP tools. Curated context injection for AI agents. P@3=86.5%, R@3=98.9%.
MIT -
Session continuity for AI coding agents — your agent picks up where it left off. Persistent memory for Claude Code, OpenAI Codex CLI & Google Gemini CLI, sharing one local db: auto context injection, compaction handover, semantic search (KO/EN/JA), error→solution recall. 100% local, zero config, zero API cost.
MIT ★ 7 -
MCP server that injects secrets into shell commands at exec time — plaintext never reaches the model context
MIT ★ 1 -
cortex-mcpgithub.com/jaswanthkumarj1234-beep/cortex-mcpupdated this year
Persistent memory for AI coding assistants. Injects context from past sessions into every LLM request.
npm MIT -
asimovs-radiogithub.com/futurespeakai/asimovs-radioupdated this year
Asimov's Radio — emotional arc orchestration through music. Standalone MCP server for session-aware musical context injection.
npm MIT -
Pre-injection context compression and session deduplication for AI coding agents: squeeze tool output, code, JSON and logs before they reach the LLM
npm (MIT OR Apache-2.0) -
claude-memory-mcpgithub.com/adeeljames/claude-memory-mcpupdated this year
MCP server that gives Claude persistent memory across sessions — reduces token usage by 70% by injecting only relevant context instead of full history.
-
sql-server-mcpgithub.com/az-coder-123/sql-server-mcpupdated this year
Model Context Protocol (MCP) server for MS SQL Server. Enables LLMs and AI agents (Cursor, Copilot, Claude) to safely explore schemas and execute read-only queries with advanced SQL injection protection.
MIT ★ 6 -
Security scanner for MCP (Model Context Protocol) servers. Detect prompt injection, secrets leaks, supply chain attacks, and vulnerabilities in MCP servers. CLI + MCP server mode.
MIT ★ 2 -
A Model Context Protocol server with machine learning engineering tools: BPE token counting and cost estimates, BM25 search, paired bootstrap and McNemar evaluation comparison, JSON Schema validation, and prompt-injection and PII scanning.
MIT -
A standalone, boundary-only secret broker — reference a secret by name, the value is injected only at the execution boundary, never inside an LLM or agent context. Free and open source.
MIT ★ 2 -
Progressive Skill Disclosure & Context Compaction Engine for Autonomous AI Coding Agents. Reduces prompt token bloat by 96.7% using JIT discovery, ephemeral schema injection, and semantic receipts.
npm Apache-2.0 ★ 3 -
A semantic data platform MCP server that composes multiple data tools with bidirectional cross-injection - tool responses automatically include critical context from other services.
Apache-2.0 ★ 10 -
cortexgithub.com/ProductionLineHQ/cortexupdated this year
Persistent memory for Claude Code. MCP server that captures decisions, context, and preferences during sessions and injects them back automatically. Local-first SQLite, quality-gated, multi-machine sync. Your AI starts every session fully informed.
MIT ★ 1 -
AI-Prompt-Guide-MCPgithub.com/Blakeem/AI-Prompt-Guide-MCPupdated this year
Deprecated. MCP server and Claude Code plugin with task driven agent orchestration workflows, guides, markdown knowledge graphs, with intelligent context injection.
MIT ★ 5 -
Local decision ledger for AI coding agents — keeps why a choice was made and which paths are dead ends, and re-injects it after context compaction. Plain Markdown, zero dependencies, works across Claude Code / Codex / Cursor / Copilot / Gemini.
MIT -
myndgithub.com/oxHive/myndactive
Persistent memory MCP server for AI coding agents. Injects project context at session start, stores architectural decisions, and recalls preferences works with Claude Code, Cursor, Windsurf, OpenCode, Kimi, and Codex.
AGPL-3.0 ★ 2 -
kiro-learngithub.com/brendangeck/kiro-learnupdated this year
Continuous learning for Kiro agent sessions on AWS. Passive tool-use capture, LLM extraction, and synchronous context injection — inspired by claude-mem, aligned with Bedrock AgentCore Memory.
npm Apache-2.0 -
Session continuity and persistent memory for Claude Code, OpenAI Codex CLI & Google Gemini CLI. Auto-injects past context on start, hands over hot files + build status across compaction, shares one local db across all three. Every feature is opt-in via `p
npm MIT -
Privacy-first Multimodal Data Refinery, Repo Context Firewall (Prompt Injection & Trojan Source Defense), Offline Vietnamese OCR, Multilingual E5 ONNX + SQLite FTS5 RAG & 9-Tool MCP Server.
MIT ★ 7 -
multi-mysql-mcp-servergithub.com/VKirill/multi-mysql-mcp-serverupdated this year
Model Context Protocol (MCP) server for MySQL & MariaDB. Multi-database, SQL injection protection, read-only mode, connection pooling, EXPLAIN ANALYZE. Works with Claude, Cursor, Windsurf, GitHub Copilot.
MIT -
AIContextBuilder (AICB): semantic code intelligence for AI coding agents on C#/.NET (dotnet). Roslyn MCP server and CLI: change impact, static analysis, code navigation (callers, implementations, dependency injection, tests, side effects, dead code) and context engineering. For Claude Code, Codex, OpenCode, Cursor; Windows desktop app included.
NOASSERTION ★ 12 -
open-agent-kitgithub.com/goondocks-co/open-agent-kitupdated this year
| Maintenance Mode | OAK automatically captures the plans, decisions, and trade-offs that usually vanish. Intelligence powered memories, context injection, autonomous agents, and more.
MIT ★ 10 -
Local-first, file-native long-term memory hub: preflight, compact context injection, citations, human-approved commits, temporal validity, dedupe/merge, retrospective reflection, write-risk flags, a derived SQLite index, per-workspace scheduled distillati
npm MIT -
A Model Context Protocol (MCP) server that exposes iTerm2 to agents for observation and non-destructive annotation. Agents can inspect sessions and decorate them (badge, title, tab color, user variables, notifications) but cannot inject keystrokes, close or spawn sessions, or otherwise alter the user's working environment.
MIT ★ 4 -
agentic-memorygithub.com/Verace-Pvt-Ltd/agentic-memoryupdated this year
6-layer persistent memory (sensory, working, episodic, semantic, long-term, procedural) with non-destructive belief revision and injection-safe context synthesis — an MCP server for Claude Code, Cursor, Antigravity, Codex CLI, and any MCP-compatible agent.
MIT ★ 1 -
Local, cross-AI memory for coding agents: Claude Code, Codex, Gemini, Antigravity, Cursor, Kimi & more share one private corpus. Hybrid search, auto-injected context, a shared skill library, a work board so sessions don't collide, and homework for ideas parked mid-task. No-API-key summarizers; runs fully local via Ollama.
Apache-2.0 ★ 5 -
mcp-injection-guardgithub.com/Hosein-Abdollahi/mcp-injection-guardupdated this year
MCP server that blocks indirect prompt injection by tracking data provenance, not text patterns.
MIT ★ 1 -
🕵️ Scans MCP servers for injection with 30+ canonical detection rules across 5 severity levels (April 2026 MCP RCE class). MIT.
MIT -
zentric-protocol-mcpgithub.com/zentricprotocol/zentricprotocolupdated this year
Prompt injection detection + PII anonymization MCP server for AI agents, with a signed audit trail. Deterministic verdicts (CLEARED/ANONYMIZED/BLOCKED), 22 signatures, 7 languages. Every request returns a SHA-256-signed audit record — GDPR Art. 30 evidenc
npm MIT -
clawguard-mcpgithub.com/joergmichno/clawguard-mcpupdated this year
MCP server for ClawGuard Shield — scan AI agent inputs for prompt injection threats
pypi MIT -
Prompt-injection defenses for Claude Code. A PreToolUse Bash hook blocks compositional credential-exfiltration shapes (secret read plus network, env dump to network, remote script to shell, reverse shells). A sanitizing MCP server wraps untrusted URLs and files in sentinels, strips invisible unicode, flags jailbreaks.
MIT ★ 4 -
MCP server for Moodle with server-side prompt-injection defenses, SSO login and Obsidian vault sync
pypi MIT -
MCP server for Moodle with SSO login, server-side prompt-injection defenses and Obsidian sync
MIT ★ 2 -
@declaw/mcp-servergithub.com/declaw-ai/mcp-serverupdated this year
MCP server for Declaw — secure AI sandboxes with network policies, PII scanning, injection defense, and audit logging
npm Apache-2.0 -
Content moderation with prompt injection detection, toxicity scoring, and PII redaction
npm MIT -
Security-focused MCP server protecting against unicode attacks, injection threats, and other AI vulnerabilities
npm Apache-2.0 -
Servidor MCP e agente LangGraph que transformam texto em notas do Obsidian, com guardrail contra prompt injection e checagem de fidelidade · Python · MCP · LangGraph
MIT -
Deterministic prompt injection classification, secret entropy detection, PII redacting, and synthetic dataset bias auditing.
MIT -
MCP server that scores text, web pages and MCP tool definitions for prompt injection and tool poisoning. Free tier, then pay per call via x402.
MIT -
MCP server: scan web pages, mails, PDFs and tool output for prompt injection before your agent reads them (17 languages).
MIT -
agent-security-mcpgithub.com/mdfifty50-boop/agent-security-mcpupdated this year
MCP server providing security scanning, prompt injection detection, secret leak detection, and agent permission auditing for AI agent workflows
npm MIT -
mcp-security-toolkitgithub.com/x0base/mcp-security-toolkitupdated this year
13 atomic MCP tools for AppSec and AI Security: recon, MCP audit, prompt-injection audit, OWASP LLM Top 10
MIT -
agent-security-mcpgithub.com/mdfifty50-boop/agent-security-mcpupdated this year
MCP server providing security scanning, prompt injection detection, secret leak detection, and agent permission auditing for AI agent workflows
-
domain-expertise-mcpgithub.com/mdfifty50-boop/domain-expertise-mcpupdated this year
MCP server that injects structured professional domain expertise into AI agents — accounting, legal, medical, Islamic finance, real estate, and more
-
@iflow-mcp/agentsafe-tooltrustgithub.com/agentsafe-ai/tooltrust-scannerupdated this year
MCP server that scans other MCP servers for prompt injection, data exfiltration, and privilege escalation. Add to your .mcp.json and let your AI agent audit its own tools.
npm MIT -
zentricprotocolgithub.com/ZentricProtocol/zentricprotocolupdated this year
Prompt injection + PII detection API and native MCP server for AI agents. 22 signatures, 7 languages, sub-25ms, GDPR Art.30 signed audit reports.
NOASSERTION -
Let Claude Code & Claude Desktop drive Firefox — safely. MCP bridge + Rust/WASM security policy that assumes the model is compromised by prompt injection.
MPL-2.0 -
Open-source engine that lets AI agents use a browser: an MCP server that shows pages as compact text, with secrets and prompt injections kept away from the model.
npm MIT ★ 1 -
MCP server that audits voice-assistant tools for prompt injection, privacy leaks and speech-unfriendly answers. Alexa+ track, Amazon Developer Hackathon 2026.
MIT -
Official MCP server for Bunker Defend. Protect Claude, Cursor, and AI agents from phishing URLs, social engineering scams, and prompt injections in real time. Zero data logging.
NOASSERTION ★ 1 -
Defensive MCP server that detects website manipulation aimed at search engines and AI assistants: hidden prompt injection, AI recommendation poisoning links, sneaky redirects, hidden spam.
npm MIT -
MCP Server for ClawGuard Shield — security scanning via Model Context Protocol.
MIT ★ 1 -
Security scanner for MCP servers, as a CLI, GitHub Action and MCP server: tool poisoning, prompt injection, secret leaks and excessive permissions, with SARIF output for CI.
npm Apache-2.0 -
MCP server to give Claude a headless browser. It includes OAuth 2.1 for security and checks for prompt injection attempts. Every render request goes through a DNS pinned egress proxy, and the project comes with a documented threat model.
MIT ★ 4 -
SOUQ // Autonomous Multi-Agent Digital Asset & Paid FastMCP Commerce Swarm
-
A GitHub MCP server that stops AI agents leaking private code when a malicious issue tells them to.
pypi MIT -
MemVault — The Shared Memory Layer for Every AI Agent You Run. Self-hosted, open-source MCP memory server with auto-injection, hybrid (BM25+vector) search, and zero-config sync — works with Codex, Claude Code, Cursor, dsh, and any Agent client.
npm MIT ★ 2 -
MCP server for IMAP mailboxes: read, search, organise and draft mail — it deliberately cannot send
MIT ★ 4 -
MCP server that sends mail over SMTP, gated behind an allowlist and a human confirmation
MIT