Dingdawg Governance
Wrap agent tool calls in a fail-closed governance layer with audit proofs and rollback before shipping autonomous workflows.
Overview
Dingdawg Governance is an MCP server for the Ship phase that enforces fail-closed policies, audit proofs, and rollback around AI agent tool use.
What is this MCP server?
- MCP-native universal governance layer for any agent toolchain
- Fail-closed execution when policies or proofs do not pass
- Audit proofs and rollback for accountable agent actions
- stdio npm package (dingdawg-governance v2.1.6)
- Optional DINGDAWG_API_KEY for paid tier; free tier available at dingdawg.com
- Server schema version 2.1.6
- stdio MCP transport via npm identifier dingdawg-governance
- Paid tier gated by DINGDAWG_API_KEY environment variable
Community signal: 1 GitHub stars.
What problem does it solve?
Autonomous agents can call tools you never meant to allow, with no durable proof of what happened when something goes wrong.
Who is it for?
Indie builders running MCP agents against production-adjacent APIs, files, or infra who need enforceable guardrails—not prompt-only pleading.
Skip if: Teams that only use chat without tool calls, or orgs that already have enterprise GRC platforms covering the same controls end to end.
What do I get? / Deliverables
Agent actions run behind explicit governance gates, leaving audit evidence and a rollback path instead of silent failures or unchecked side effects.
- Registered stdio MCP governance server in agent config
- Fail-closed tool execution with audit proof trail
- Rollback path when governed actions violate policy
Recommended MCP Servers
Journey fit
How it compares
Policy and audit MCP layer, not a single-task skill or generic LLM wrapper.
Common Questions / FAQ
Who is Dingdawg Governance for?
Solo and small-team builders shipping agent automations who need MCP-native policy enforcement, proofs, and rollback without building governance from scratch.
When should I use Dingdawg Governance?
When your agent can invoke tools that change state—deploy, delete, charge, or exfiltrate—and you want fail-closed checks before those calls execute in production.
How do I add Dingdawg Governance to my agent?
Install the npm package dingdawg-governance, register it as an stdio MCP server in Claude Code or Cursor, set DINGDAWG_API_KEY if using paid tier, and point your agent config at the server entry.