claude-flow
Verified · 20 days agoAI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
THOR
Verified · todayLocal memory for coding agents. A rule with a runnable proof can refuse a wrong command.
Install
claude mcp add thor -- docker run -i --rm ghcr.io/nworks3d/thor-mcp:2.0.0
Our take
THOR is positioned as local memory for coding agents with a rule system that can block commands when it has executable proof they are wrong. It targets developers who want stronger local safeguards and memory in agent-assisted coding environments. The idea is interesting but quite ambitious relative to the small maturity signals, so it reads as experimental infrastructure that warrants close inspection before relying on it heavily.
Verification record
- today
- 3
- today
- no
- GPL-3.0
- 2026-08-15
More in ai
browse all →tldraw
Verified · todayDraw and visually collaborate with your agents on tldraw's canvas.
Codebase Memory
Verified · 17 days agoCodebase knowledge graph for AI agents — 159 languages, sub-ms queries, 99% fewer tokens.
OpenMetadata
Verified · 12 days agoOfficial OpenMetadata MCP: governed context and business semantics for AI assistants and agents.
mobilerun
Verified · yesterdayControl real Android and iOS devices with LLM agents — tap, swipe, type, automate flows.
PraisonAI
Verified · 16 days agoAI Agents Framework with Self Reflection and MCP support