AI & Agents MCP Servers for DeepSeek Harness

Vantio Gate — Policy Latch dry-run evaluate for agents. No live enforce from MCP.

Session handoff for Claude Code: list sessions and build handoff docs another agent continues from.

Write Velaris, audit what it can touch, and run it under an effect budget.

Zero-dependency verifiable-claims MCP: validate safely, verify with allow_execution, self-test.

A post-mortem succession protocol for AI agents. Black Box, handoff, time-locks.

Submit AI inference to XFuel, pay per task (USDC via x402 or TFUEL), fetch/verify ZK proofs.

Open-source AI context firewall for governed model, retrieval, memory, and tool access.

A2A v1.0.1 agents as MCP 2026-07-28 tools, with the tasks extension

Claude Code MCP plugin and wakeable worker for direct local messaging with Pi, Codex, and Claude coding-agent sessions.

MCP proxy server that allows MCP clients to communicate with agent servers (Mastra, LangGraph, etc.)

MCP server for AgentForge — trust audit (check_trust, evaluate_policy, list_trusted, recommend) plus action tools (discover

Explainable ALLOW/WARN/BLOCK risk decisions for AI agent actions before execution, signed as OAA.

Tamper-evident, SHA-256 hash-chained audit trail and observability for AI agents.

Quality verification for AI agents and MCP servers. 6-axis scoring, adversarial probes.

Reference MCP server: a demo bakery agents can query for hours, menu, availability, and orders.

Document reasoning checks: a helicopter view of claims and assumptions before handoff.

MCP server for the Codabench REST API — drives a full ML-benchmark participant workflow.

CognOS trust scoring (C=p·(1-Ue-Ua)) and session trace storage as MCP tools.

Compare one prompt across a multi-model panel side by side, with public verification receipts.

Decompose text into classified semantic units. Authority, risk, attention. No LLM.

Polyphonic meritocratic ensemble for Universal IDEs