Dingdawg Loop

dingdawg/loop-protocol

This wraps scheduled agent actions with governance gates that block high-risk operations until policy thresholds are met. You register loops with cron schedules and governance rules, then execute them through MCP tools like register_loop, execute_loop, and loop_audit. Every execution gets signed with an Ed25519 receipt for audit trails. It categorizes operations as destructive, financial, sensitive, or bulk, and requires elevated approval for risky actions before they run. Works with Claude Desktop, Cursor, CrewAI, and LangGraph. Reach for this when you need scheduled AI agents that won't accidentally delete production data or send mass emails without review.

tools

6

auth

Secret needed

transport

STDIO

env var

1

last updated

May 26, 2026

Sponsors

Promote your product hereReach 125k+ visitors monthlyPromote your product hereReach 125k+ visitors monthlyAgent CaffeineSponsoredClose your lid, come back to finished workSnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthlyWaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthlyknobsSponsoredSee your site on any phone, any day, in any language

npx dingdawg-loop

{
  "mcpServers": {
    "dingdawg-loop": {
      "command": "npx",
      "args": ["dingdawg-loop"]
    }
  }
}