AI & Agents MCP servers

Persistent memory MCP server for AI coding assistants. Chinese-first and fully local.

Auditable memory for coding agents: evidence chains, supersession history, keyless embedded SQLite

Audit and generate agent-native product surfaces

Cost-parity bench for agent loops: $/success and time across model mixes, with verifiable receipts.

Long-term memory for AI agents: local-first PKM with provenance and hybrid search.

Delegate heavy tasks from Claude Code to the Antigravity CLI (Gemini) with quota-aware failover.

Batch-dispatch coding tasks to Google Jules: 50 parallel sessions, plan approval, PR collection.

Dynamic pod spawner & proxy for ephemeral AI agent workspaces on Kubernetes without CRDs

Checks agent traces for evidence-readiness: verifies evidence is present, not that it is true.

Verify, discover, and rate AI agents before transacting.

Identity, memory, trust, and swarm MCP tools for AI agents. Spark trial at lithtrix.ai.

Lossless context compression: 2-8x fewer tokens, byte-exact recovery, search inside payloads

A neutral memory quality-gate MCP server: save_memory, recall_memory, forget_memory over stdio.

Append-only event log for cross-agent context/memory sharing via 3 MCP tools.

Inspect and score any MCP server 0-100 on publishability, schema and protocol before publish.

Discover and publish AI agents via DNS using SVCB records (RFC 9460)

Local-first, source-cited memory for AI agents with proposal review, failure memory, and MCP tools.

Bring your GFT (Git for Thought) white-box thinking map to any AI agent. 把你的思维脉络接给任何 agent。

Drive the Cursor agent from Telegram, Discord, or GitHub — with images and voice notes.

Governed memory for AI agents (Paramind AI): conflict-checked writes, time-travel reads, ACL, audit.

Hypothesis-driven problem solving for AI agents: probe, falsify, escalate.

Local agent spend-policy checks. No funds, keys, payment authority, or network access.

x402 service-quality oracle: check a service's paid-probe quality tier and score before paying it.

LoopPause MCP server — pause agent execution and route human approval requests

Agent Relay MCP: invite-based channels, agent contracts and per-side human approvals.

Local-first, zero-knowledge AI memory across Claude, Cursor, and every MCP client. 0 RTT reads.

Persistent, expiring, importance-graded sticky notes for Claude Code.

Persistent cross-session memory for OpenCode and MCP coding agents, powered by ZeroDB.