"An open-source penetration testing framework" matching MCP connectors:

Specialist tools for any job — a lead, an image, a song, live data, and more.

Prompt evals over MCP: run a prompt on your dataset, score each output 1-5 with an LLM judge.

Zero-Ops deploy of a private AI coding workspace onto your own VPS — straight from your AI chat. Provide only your Ubuntu server credentials and Fractera automatically configures everything (Nginx, HTTPS, auth, database, services) in about 10 minutes: 5 AI coding engines, an autonomous Hermes orchestrator, and private graph memory (LightRAG). No terminal, no DevOps. IP-first and free; a custom domain with HTTPS is an optional later step.

clariBI is an AI-powered business intelligence platform built for small and mid-sized SaaS, ecommerce, and operations teams. Instead of writing SQL or building dashboards manually, users ask questions in plain English ("what's our CAC by channel last quarter?") and the AI engine pulls the data from Stripe, HubSpot, Google Analytics, ad platforms, Jira, and 30+ other sources via the open Model Context Protocol. Dashboards generate themselves on source connect.

Open-source AI accounting skills verified by licensed accountants (tax, VAT, payroll).

Free Memory-First AI Operator course as an MCP server. Bundled curriculum, no account needed.

LibreTranslate MCP — open-source machine translation (BYO endpoint)

Discover open-source AI agents and get source-owned installation instructions through MCP.

Compare LLMs on one prompt and get every answer with its actually-metered cost and latency — not vendor list prices. Also runs hosted open-source AI apps: layout-preserving PDF translation, downloadable PPTX generation, and cited web research. Nine tools over Streamable HTTP. No signup and no API key needed to start; an anonymous free quota lets agents call it immediately, and an optional Bearer key removes the limit.

Check if an AI model is deprecated, retiring, or silently changed price or context window.

Prioritize an existing Italian B2B lead list into up to 250 explained JSON decisions.

Check whether an LLM model id is deprecated, retiring, or retired, and what to migrate to.

Courses that argue back: learn adaptively, build real work, then defend it to an AI stakeholder.

Adversarial verification for AI agents - pay an independent skeptic per verdict in USDC via x402.

Deterministic prompt-injection detector; signed, offline-verifiable verdicts. Not an LLM.

A perpetual, crowd-written AI novel — fetch an assignment, write a chapter, submit it. CC0.

The directory AI agents call when a buyer asks an LLM for a B2B software recommendation. Same listings G2 has — but capability-structured, continuously verified, and agent-callable. One listing visible to both audiences.

Five-to-twelve-cent source-attributed public URL and domain result packs for AI agents.

Get a real human to verify, decide, or improve an AI agent's work.

Manage and optimize SiteSpeakAI chatbots from an AI agent in plain English: read analytics, tune prompts and model, manage knowledge sources and suggested messages, enable lead capture and human handoff, and triage the inbox.