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.
-
A local harness for Claude Code and Codex: budgeted test runs with a stop rule, a guard for the tests you already have, and a token tree for every subagent.
MIT -
Local-first context ingestion and retrieval for AI tools. SQLite + embeddings + MCP server for Cursor & Claude.
AGPL-3.0 ★ 43 -
Regression-Aware Context Compaction & Witness Replay for Coding Agents (CEGAR-Harness)
MIT -
Shared context for AI agents — structured harnesses from conversations, feedback, and decisions, queryable via MCP
npm MIT -
Harness-agnostic inter-agent proxy and Model Context Protocol (MCP) server for AI coding agents
MIT ★ 2 -
Local-first code knowledge graph and context harness for coding agents. MCP + DeepSeek Harness (dsh) plugin.
Apache-2.0 ★ 30 -
DeepSeek Harness (DSH) plugin that publishes the running Harness as a Model Context Protocol (MCP) server for other agent harnesses.
npm MIT ★ 6 -
The agentic meta-harness — freeze the model, evolve the harness. An open runtime that routes each query to the cost-optimal model, evolves its own harness (planner/context/reviewer/retry/tool/memory/score policy) and autonomously repairs code, then orches
npm MIT -
Open-source IDE context harness and persistent memory engine for AI coding agents
npm MIT -
A TypeScript-based Model Context Protocol (MCP) server for inter-session terminal orchestration, polyglot harness multiplexing, and AI agent control planes.
GPL-3.0 -
CoAligne for DeepSeek Harness: shared team project context over MCP, plus the coaligne-workflow skill
npm NOASSERTION ★ 2 -
TypeSafe Jev for DeepSeek Harness, the Model Context Protocol, and plain Node: typed judgments instead of prose, offline by default.
npm Apache-2.0 ★ 105 -
DeepSeek Harness bundle that exposes Model Context Protocol (MCP) servers as native tools, with a settings UI, live status, auto-reconnect, and backup/restore via local files or WebDAV.
npm MIT -
A GUI switch for the REA MCP server in DeepSeek Harness — turn its 229 mcp__rea__* tools on or off, so unrelated chats stop paying their context cost.
MIT ★ 1 -
Micro ECF is the open local policy and context harness for coding agents: dependency-free CLI plus MCP server, durable ECF.md, source maps, budgets, approvals, and Agent OS exports. No cloud or wallet.
Apache-2.0 -
MCP switch plugin for DeepSeek Harness: MCP servers stay disconnected by default (no context cost), toggle them from agent tools (mcp_switch_*) or a small Settings panel.
MIT ★ 2 -
Structural long-term memory for DeepSeek Harness: supersede/decay/entity-state store, 15 MCP memory tools, automatic recall, bounded context band
npm MIT ★ 3 -
Windows desktop UI context picker for AI agents: select windows, UI elements, or screen regions with hover highlighting, UI Automation, screenshots, local OCR, and user-approved MCP output. Works with OpenAI Codex, DeepSeek Harness, and other MCP-compatible clients.
MIT ★ 2 -
dsh-memory-protocolactive
Memory protocol enforcement plugin for DeepSeek Harness — forces memory_weave before tool calls, auto-ingests session turns, injects memory context into agent steps. Bridges the opencode-memory MCP server with a hard protocol gate.
npm MIT -
The context quality layer for AI agents — memory that checks itself: lifecycle governance, calibrated confidence, and staged claim gates. Local-first, MCP 13 tools, DeepSeek Harness plugin.
npm Apache-2.0 ★ 12 -
This is the CtrlAsistent MCP (Model Context Protocol) server, which works with any MCP client: - llama.cpp - OpenCode - DSH Harness - Pi Harness - Claude Desktop - and others
MIT -
the-metis-fca-handbook-ai-harness-mcpno update data
# Metis FCA Handbook AI Harness — MCP Server An MCP (Model Context Protocol) server that integrates the Metis FCA Handbook AI Harness into Claude Desktop and other MCP-compatible agents. **Tool name:** `evaluate_fca_handbook_applicability` Use this tool to analyze FCA Handbook applicability for a product, service, or firm. Returns structured compliance reasoning with verbatim handbook citations. ## Why choose The Metis FCA Handbook AI Harness? - **Efficiency** Multiplies effectiveness of compliance advice. Saves £80+ in token fees per Harness run. - **Deals with "Hard Problem"**, in which LLMs' token incentives prioritise training data so results are unreliable. - **Verbatim citations** Quotes verbatim entries from the FCA Handbook. Other AI systems struggle to do this. - **Matches real-world need** You do not have to start off knowing which sections you are looking for. Carries out structured searches across all 10,000+ FCA Handbook entries. - **Secure Design** Harness compartmentalisation, one-shot structure, and statelessness fits natively with emerging AI agent security standards such as OWASP Top 10 for Agentic Applications 2026. - **AI Accessible** Integrates easily with AI workflows and agents. Your LLM can use it as a tool. ## Design Principles The Harness is built on proven principles: - **Stateless** — Each request is independent; no session coupling - **One-shot** — Complete analysis in a single call; no multi-turn state - **Clear contract** — Explicit input/output schemas for easy integration - **Hosted** — Single canonical source; no version drift or stale data ## Use Cases ### Compliance AI Platforms Embed FCA reasoning as a service within your compliance platform. Users ask natural-language questions; your platform calls the Harness and presents structured reasoning. ### AI Agent Workflows Agents building compliance workflows can include FCA Handbook reasoning as a composed tool—no external API calls, just MCP configuration. ### Compliance Review Automation Integrate into document review or due-diligence pipelines. Automatically screen new rules against FCA applicability. ## For Regulatory Screening FS firms screening MCP servers will find: - **Simple model:** Stateless, no hidden state, no background jobs - **Clear contract:** Explicit input schema, output schema, error modes - **Transparent updates:** Version pinning; no automatic upgrades - **Single point of failure:** If the Harness is down, it is obvious; no cascading config issues - **Audit-friendly:** All calls logged centrally, not distributed
smithery -
cligithub.com/prjct-app/cliactive
The agentic harness for AI coding agents — intelligence is rented, the harness is owned. Runs the full work cycle: intent briefs, bounded RAG context from real code intelligence, persistent memory, preventive guardrails, and performance evals. One CLI + MCP server, SQLite inside. Claude Code, Codex, Gemini CLI, Cursor & OpenCode.
npm MIT ★ 1 -
⚡ Fast, local, provenance-aware Code-Context MCP Server & DeepSeek Cordis Harness. Compresses 500k+ tokens by 99.9% via AST elision, PageRank & Progressive Zoom-in for Claude Code, Cursor, Antigravity.
pypi MIT -
Model Context Protocol server exposing structured HR policy tools, RAG context retrieval, and an automated evaluation harness.
npm MIT -
Production-grade context engine for AI agents — hybrid RAG, reranking, token-aware assembly, MCP server, eval harness
-
@ricky.instrument/spells-mcpactive
Model Context Protocol (MCP) server and prototyping harness for Instrument Wizard Spells
npm UNLICENSED -
toolbeltgithub.com/aragaobruno/toolbeltupdated this year
SRE-grade AI Agent Toolchain & Skill Harness. Optimized for context efficiency, robust rate-limiting, and deterministic tool execution.
MIT ★ 2 -
A durable, harness-agnostic memory + context layer for coding agents and MCP clients.
npm Apache-2.0 -
Full-context Google Workspace MCP server: 73 tools, cursor envelopes, staged writes, harness-agnostic
MIT ★ 2 -
Context compaction for agent harnesses: a decision model scores transcript spans and tombstones the ones that stop mattering. Includes an MCP restore server.
MIT -
The Model Context Protocol server for @kitn.ai/ui: component reference, scaffolder, theme and debug tools for any AI coding harness.
npm MIT -
posthog-mcp-minigithub.com/josephruocco/posthog-mcp-miniupdated this year
An MCP server that serves PostHog's JS/Web SDK docs as agent-optimized context, with an eval harness that measures it against a naive baseline.
MIT -
Context-APIgithub.com/EditorialOS/Context-APIupdated this year
Memory and brand-context layer for AI agent systems. Multi-tenant MCP server — 39 tools auto-generated from a live OpenAPI spec. PostgreSQL, per-workspace isolation, scored eval harness.
NOASSERTION -
skillsgithub.com/sattva2020/skillsupdated this year
Agent skills and MCP servers for AI-assisted development: harness audit (context, tools, memory, evals, guardrails), Dokploy API MCP
★ 3 -
Model Context Protocol (MCP) server suite and verification harness for RFP compliance, tool auditing, and x402 Base USDC metering.
-
Bridge AI assistants to the Unity Editor over MCP (Model Context Protocol). MCP for Unity v10.3.0 plus ZCode and DeepSeek Harness client support.
Apache-2.0 -
forgetmenotgithub.com/djroxx2000/forgetmenotupdated this year
A local harness for mem0 to persist memories from sources such as LLM context windows to durable stores (Qdrant - Vectors, Neo4j - Knowledge Graph, Postgres - Archival) via MCP
MIT ★ 1 -
Measure what context compression actually costs an LLM agent: where a session's tokens go, and whether a rewrite pays for the prompt cache it invalidates. Rust crate, MCP server and benchmark harness.
npm MIT ★ 1 -
beebjit-mcpgithub.com/acscpt/beebjit-mcpupdated this year
Model Context Protocol (MCP) integration for beebjit enabling structured context exchange between the BBC Micro emulator beebjit and AI tools. Also a Python harness for direct CI/CD integration or other testing pipelines.
pypi MIT ★ 1 -
Full AI context layer over MCP — tree-sitter code-map, document RAG (PDF/Office/HTML/email + OCR + reranker), shared agent memory, on-demand web crawl, git history + blame + per-symbol diff. 300+ languages, 10+ coding-agent harnesses, content-addressed Fjall + LanceDB.
pypi MIT -
RushDB for AI coding agents: Claude Code, Codex, Gemini CLI, Qwen Code, Antigravity, Pi and DeepSeek Harness. MCP tools, skills and optional automatic context.
Apache-2.0 -
Local-first code context CLI + MCP server: discover, index, and search repos; generate task-aware context packs, code maps, agent harnesses, and PR-review/merge-readiness gates for coding agents and reviewers.
npm MIT ★ 3 -
x16-emulator-MCPgithub.com/acscpt/x16-emulator-MCPupdated this year
Model Context Protocol (MCP) integration for the x16-emulator enabling structured context exchange between the x16-emulator and AI tools and a Python harness for direct communication and integration with CI pipelines.
pypi MIT -
kakeibogithub.com/siash1/kakeiboupdated this year
A personal finance agent built from scratch on the raw Gemini API: hand-written tool-calling loop, double-entry PostgreSQL ledger, context caching, MCP server, 46-task eval harness, prompt-injection suite.
MIT -
Long-term memory MCP server for OpenCode and other harnesses — stores preferences, lessons, and usage history in a single local SQLite file (100% local, no external API) so the AI can "remember and adapt" to the user through context-based learning.
npm MIT ★ 3 -
MCP server that hands bulk, read-heavy work from Claude Code to a local DeepSeek Harness agent — it opens files itself, so their text never enters your context. Well suited to 1C:Enterprise (BSL).
MIT ★ 2 -
AriaWave METACOGNITION™ (v1.0): Self-auditing recursive agent harness, Epistemic Confidence & Entropy Firewall (ECEF), Read-Live/Write-Draft MCP Cleanroom Gate (HMAC-SHA256), Anti-Amnesia 180KB Rolling Context Slimmer, and Pre-Flight Soft-404 Verifier.
NOASSERTION -
Your agent should already know what you're looking at. A local context harness for the browser: it reads the page and how you got there, and hands it to Claude Code, Cursor or any MCP agent.
MIT ★ 1 -
MCPGoatgithub.com/SabyasachiDhal/MCPGoatupdated this year
Deliberately vulnerable MCP server (aka MCP Goat) for security training — 26 challenges across 4 difficulty levels (incl. a secure reference), a victim-agent harness, and one-command Docker deploy. Practice penetration testing against the Model Context Protocol.
MIT ★ 12 -
asoftimegithub.com/amatayomosley-web/asofupdated this year
Temporal-awareness hook for tool-using LLMs. Surfaces freshness verdicts on in-context data so stale information doesn't poison fresh reasoning. Adapters for Claude Code, Antigravity, generic harnesses (LangGraph, CrewAI, raw SDK pipelines), and an MCP server for any MCP-compatible client.
pypi MIT -
A production-style multi-agent system that generates context-aware briefs. Built on MCP + A2A, with persistent memory (Chroma), dynamic tool routing, self-critique, distributed tracing (Arize Phoenix), an LLM-as-judge eval harness, and Redis caching.
★ 1 -
Codebase Q&A with (path:line) citations — the capstone of a from-scratch AI-engineering series. Built in 8 eval-gated stages (chat → RAG → agents → hardening → production), provider-agnostic and offline-first, with 4 extensions (MCP, agent harness, context management, local models).
MIT ★ 1 -
A sub-harness (both an MCP server and an MCP client). Delegates documentation and source code analysis, as well as interactions with related context-providing MCP servers, to a local or inexpensive OpenAI-compatible LLM. Drastically reduces token usage and context size for coding agents on the top-tier LLM.
Apache-2.0 ★ 3 -
Graph-Augmented Code Intelligence RAG engine pairing deterministic AST call graphs with dense vector search to boost direct-callee Recall@8 from 0.0% to 53.3% (+53.3 pp) under a 5.8ms p95 latency bound. Powered by a QLoRA fine-tuned Qwen2.5 adapter (70.02 CodeBLEU), quality verification harness, and production Model Context Protocol (MCP) server.
MIT -
Reference Model Context Protocol (MCP) server for the Agentic Code Orchestration Protocol (ACOP). Exposes the ACOP coordination write surface — claim, complete, create work item, accept handoff, post blackboard — over stdio JSON-RPC, so MCP-capable agents (Claude Code, Codex, custom harnesses) can drive coordination state via tool calls. Coordination authority is delegated to a pluggable backend adapter; the shipped HTTP adapter targets any ACOP-compatible coordination middleware via a configurable base URI.
nuget Apache-2.0 -
Launcher for the XuanLing result-only harness bridge (MCP stdio).
npm MIT -
@abranzhrl1212/agent-harnessactive
High-performance native Rust MCP server for AI code assistants
npm MIT -
Cross-harness MCP connector for coding agents
npm MIT ★ 1 -
AIM — AI Maestro for the Harness platform. Your personal Harness Architect, powered by the Harness Adoption MCP server.
npm Apache-2.0