mcp — independent software & tools
-
patient-zero
Supply-chain attack scanner for the agent era. Triage in 30s with `npx patient-zero`, block malicious installs before postinstall runs, or drop into CI as a GitHub Action. Covers npm + Python + MCP agent configs. Free, MIT, no signup, no telemetry.
-
boardstate
Your dashboard is data. Any AI can build it; any human can edit it. A protocol + runtime for agent-composable dashboards — layout-as-data, one guarded control plane, sandboxed agent-authored widgets, MCP server included.
-
Agent-Worlds
-
geode
Geode provides agents remote access to your Obsidian vault.
-
cast
Minimal terminal coding agent with OpenAI-compatible routing. Works with any provider — OpenRouter, Ollama, vLLM, LiteLLM, or your own.
-
cast
Minimal terminal coding agent with OpenAI-compatible routing. Works with any provider — OpenRouter, Ollama, vLLM, LiteLLM, or your own.
-
context-optimizer
Reduce Claude token consumption by using local manifests to index and summarize project files for efficient context management.
-
llm-wiki-patterns
🚀 LLM Wiki Graph Studio 2026 - Smart Research, Lint & Content Pipelines
-
mcp-remote
The bridge for remote MCP — seamless OAuth, resilient auth recovery, production-grade reliability
-
mcp-webasyst-codex-toolkit
Unofficial Codex-First MCP for Webasyst 2026 – Apps, Plugins & UI Audit Tools
-
session-recall
Shared memory for Claude Code and Codex. Your agent recalls decisions from weeks ago — including what the other engine worked out. Local semantic search over both histories, as MCP tools.
-
veilbrowser
Stealth browser for AI agents — real Chrome over raw CDP, no Playwright/Puppeteer. TypeScript + MCP-native. Passes sannysoft 57/57, bypasses Cloudflare.
-
openmarket-ai
OpenMarket.ai — Agent-to-agent marketplace on Hedera. AI agents buy and sell services. Auto-payments via HBAR. MCP, SDK, LangChain, CrewAI.
-
pocketcast-cli-commander
Pocket Casts AI Search 2026: Smart Episode, Show & Metadata Explorer
-
skill-manager
Manage and sync AI skills and MCP servers for coding tools with Git-based updates in one desktop interface.
-
ctxone
Persistent, searchable, accountable memory for AI agents — the shared team memory layer. Eliminate context anxiety.
-
hspace
Ship to one, not to all. Targeted sharing for AI-built HTML/Markdown: one link + password, view receipts, instant revoke, per-recipient passwords, versioned. Open source, edge-native.
-
toksave
A zero-config CLI that securely installs and wires token-saving tools into AI coding agents.
-
unityxclaude
Control Unity Editor with Claude via MCP to create scenes, edit components, manage assets, run builds, and execute C# from chat
-
shelf
Local tool library for macOS — keep the tools you build, launch them in a click, and share them honestly with your AI agents over MCP. Local-first, no account, MIT.
-
clauderabbit
ClaudeRabbit is a free, open-source security product protecting the open-source community from malware. We clone any public GitHub repo into an isolated sandbox, run it, and return one honest safety score.
-
aileron
Tamper-evident flight recorder for AI agents - hash-chained, signed audit log of every tool call, with MCP proxy policy enforcement and anomaly detection.
-
aipermission
Local-first permission gateway for AI agents: connector-based SSH, Postgres, and Redis access with scoped tokens, human approval, audit logs, and encrypted local credentials.
-
leadace
AI sales rep for Claude Code. Finds prospects, researches each company, writes and sends outbound email from your own inbox, and learns from every rejection. Self-hostable (Cloudflare + Supabase).
-
LexonGraph
LexonGraph is a log-structured semantic index that stores embeddings as immutable blocks in navigable vector graph. It supports local and cloud backends, fast centroid-guided search, and efficient period rebuilds, enabling LLMs to retrieve precise context from large archival corpora.
-
text-metrics-analyzer
🚀 PretextPlugin 2026 – Deep MCP Tools for DOM-Free Text Layout & Measurement
-
roadkeep
Keeps a project's roadmap, changelog, improvements and strategy files to one format - a schema at the point of insertion, not a linter after the fact. Python CLI + Claude Code plugin.
-
alertint-agent
Read-only Alertmanager intelligence agent — correlates alerts, triages incidents with LLMs, and exposes findings via MCP to AI coding agents
-
renforge-mcp
Free, open-source MCP server for Ren'Py visual-novel development — give AI agents full control of a Ren'Py project (analyze, write, run, see, test).
-
Obsidian-vault-api
Obsidian plugin that exposes your vault as a local MCP server for Claude Desktop app
-
grounded-reasoning
Zero-token, precision-guaranteed relation-algebra verifier for LLM/agent multi-hop reasoning
-
AlfredOS
An open-source, self-hostable, multi-user, multi-persona, security-hardened agentic OS.
-
huqan
Local-first, deterministic verification layer for AI claims, memory writes, and risky actions.
-
mymeal
Self-hosted recipes, AI meal planning, pantry & smart shopping lists — Home Assistant native (add-on + HACS integration, MCP, Assist voice)
-
webasyst-ai-workbench
Webasyst MCP Server 2026: AI Tools for Apps, Plugins & Themes
-
terse-sharp
Roslyn-powered MCP server for .NET — your coding agent stops reading whole C# files. 13-30x fewer tokens: semantic navigation, compile-gated edits, solution-wide rename, and type-checked XAML bindings for WPF/Avalonia/WinUI/MAUI. No IDE, no licence, one install.
-
agentweb
Turn websites into fast CLI commands that any coding agent can use
-
larapilot
Laravel spec-driven workflow
-
AgentAlertsSkill
Agent skills for Codex/Claude to set up automations that send iPhone Live Activity, widget, notification, and Control Widget updates to your iPhone.
-
apexjs
Apex JS — HTML-first apps, AI-native APIs. The full-stack, AI-native meta-framework for Alpine.js: file routing, SSR, islands, and every typed route is also an MCP tool.
-
ani
A tiny personal agent: DeepSeek brain, QQ/Telegram body, pi soul. Zero-dependency TypeScript on Node 24.
-
edgeverdict
A code review gate that verifies changes by executing tests, not by judging diffs. LLM proposes; a deterministic gate decides.
-
kotlin-lib-mcp
MCP server exposing the sources, public API and KDoc of Maven-published Kotlin/Java libraries
-
oracle-forms-mcp
MCP server that exposes Oracle Forms modules (.fmb/.mmb/.pll/.olb) — blocks, triggers, PL/SQL, object XML — to AI assistants for legacy modernization and code review.
-
aegis
The meta-harness that brings real agentic collaboration as a first-class citizen
-
shipwright
The open-source autonomous delivery agent for Claude Code.
-
nahuali
Self-inspecting, auditable memory for AI agents: it surfaces the evidence, provenance, and health behind each recall so callers can see which memory to trust. Optional Ed25519-signed tamper-evident ledger. Local-first, Rust.
-
Chaos-MCP
Chaos-MCP is an MCP stdio server that runs isolated mutation testing against a target codebase to expose gaps in its test suite. It wraps four language engines — StrykerJS (TS/JS), cosmic-ray (Python), cargo-mutants (Rust), Infection (PHP) — and exposes three tools: audit_code_resilience, triage_test_coverage, and estimate_audit.
-
glossa
Make ChatGPT your local coding agent
-
revops-skills
Automate Salesforce RevOps tasks with Claude agent skills designed for pipeline hygiene, forecast reconciliation, and deal pressure testing.
-
smolanalytics
open-source web + product analytics in one Go binary. ask your data in plain english, answers provably match the dashboard (a CI test enforces it). your coding agent instruments your app over MCP. self-host free.
-
claude-code-reverse-engineering
Explore reverse engineering specs for Claude Code CLI, with implementation-ready design docs from source map analysis of the npm package
-
exomem
Self-hosted MCP server that makes your Obsidian/markdown vault searchable — text, PDFs, Office docs, images, audio — from any MCP client. Hybrid retrieval over a typed, governed corpus, sub-second at 50k notes; your files stay plain markdown.
-
claudinho
⚽ 2026 World Cup live scores in your terminal — MCP server, CLI & statusline for Claude Code and Cursor. No API keys. Not affiliated with FIFA or Anthropic.
-
FixMap
Local-first repo maps for coding agents—ranked files, test routes, risks, CLI/MCP/GitHub Action, and public GitHub URLs.
-
Actionables
Local execution coordination for OpenAI Codex. Turn technical findings into scoped, evidence-backed tasks with MCP integration and validation.
-
ashlr-hub
Local-first command center for agentic engineers — the CLI + Raycast hub and front door to the Ashlr dev-tool ecosystem.
-
attestral
Open-source security scanner for MCP servers and AI agents: finds prompt injection, tool poisoning, and excessive agency, and models the cloud your agents can reach. pip install attestral.
-
icode-skill
A 6-step coding workflow for Claude Code — Plan → Review → Finalize → Code → Deep Check → Audit. Run all at once or step-by-step, switch models between steps. Claude Code 六步编码工作流：计划 → 审查 → 定稿 → 编码 → 复检 → 终审，一步到位或分步执行，步骤间可切换模型。
-
3ngram
Persistent, typed memory for AI agents — open-source MCP server, REST API, SDK, CLI, worker, and self-host backend.