claude-flow
Verified · todayAI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
penholder
Synced · awaiting checkPreventive human-approval write-gate for AI agents: writes commit only after a human approves.
Install
claude mcp add --transport http penholder https://api.penholder.ai/mcp
Our take
This server adds a human-approval gate before agent write actions are finalized, with an emphasis on committing changes only after review. It is for teams running coding or automation agents in environments where auditability and change control matter more than full autonomy. The concept is clear and defensible, though narrow by design, and the listing does not offer enough maintenance or adoption signals to judge how battle-tested it is.
Verification record
- today
- 2026-08-17
More in ai
browse all →tldraw
Verified · 2 days agoDraw and visually collaborate with your agents on tldraw's canvas.
Codebase Memory
Verified · 19 days agoCodebase knowledge graph for AI agents — 159 languages, sub-ms queries, 99% fewer tokens.
OpenMetadata
Verified · 14 days agoOfficial OpenMetadata MCP: governed context and business semantics for AI assistants and agents.
mobilerun
Verified · 3 days agoControl real Android and iOS devices with LLM agents — tap, swipe, type, automate flows.
PraisonAI
Verified · 18 days agoAI Agents Framework with Self Reflection and MCP support