The Agent Control Plane - Default-Deny Security for MCP Servers
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"mcp-server-nucleus","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# 🧠 Nucleus Sovereign OS [](https://badge.fury.io/py/mcp-server-nucleus) [](https://nucleusos.dev) [](https://youtu.be/jI8TUpfjS1A) [](https://opensource.org/licenses/MIT) > **The Operating System for AI Agents** — Persistent Operational Memory, Swarm Orchestration, and Local-First Sovereignty. Nucleus is the **Recursive Aggregator** that gives your AI agents a persistent brain (`.brain/`) and a file system. It turns stateless chatbots into stateful **Sovereign Agents**. ### Context vs. Control Claude's `CLAUDE.md` provides **static context**. Nucleus provides **active control**. | Feature | CLAUDE.md / .cursorrules | Nucleus (Agent Control Plane) | | :--- | :--- | :--- | | **State** | Static (read-only text) | **Dynamic** (Stateful DB, Event Ledger) | | **Memory** | Session-bound (forgotten on close) | **Persistent** (Project-bound, recallable) | | **Security** | None (Prompt injection risk) | **Enforced** (Auth boundary, Default Deny) | | **Tools** | Suggestions only | **Orchestrated Execution** (DAGs) | | **Audit** | None | **Full Decision Trail** (Who/Why/When) | ## ✨ Governance Features (The Moat) - **Default Deny Security** — All mounted servers start with NO network/filesystem access. - **Explicit Consent** — You approve every command. No silent execution. - **Isolation Boundaries** — Tools cannot see each other or the full chat history. - **Auth Firewall** — Tokens are stored in Nucleus (Host), never passed to agents. - **Event Ledger** — Immutable audit trail of every agent decision (`DecisionMade`). - **Decision Provenance** — v0.6.0 DSoR: Full audit trail with context hashing. - **IPC Security** — Per-request auth tokens prevent socket impersona…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.