AcademyMCP serversOperations

DashClaw

Operations and IT teams that run AI agents unattended and need a human to approve risky or costly actions.

  • Listedour rating
  • 312gitHub stars
  • MITlicence
  • todaylast update
claude mcp add dashclaw -e DASHCLAW_URL=https://your-dashclaw.vercel.app -e DASHCLAW_API_KEY=oc_live_xxx -- npx @dashclaw/mcp-server
README.md

Loading the file...

What it does

DashClaw sits between an AI agent and the actions it wants to take. It checks each action against your organisation's policy, asks a person for approval when needed, and records every decision in a ledger. If a check cannot be completed, the action does not run.

Use cases

  1. 01Require approval before an agent deletes data
  2. 02Block agent actions that break company policy
  3. 03Keep a decision log for internal audits

Questions about
DashClaw.

What is DashClaw used for?

Operations and IT teams that run AI agents unattended and need a human to approve risky or costly actions. DashClaw sits between an AI agent and the actions it wants to take. It checks each action against your organisation's policy, asks a person for approval when needed, and records every decision in a ledger. If a check cannot be completed, the action does not run.

How do I install DashClaw?

Run this in your terminal: claude mcp add dashclaw -e DASHCLAW_URL=https://your-dashclaw.vercel.app -e DASHCLAW_API_KEY=oc_live_xxx -- npx @dashclaw/mcp-server

Is DashClaw open source?

Yes. The code is on GitHub (ucsandman/DashClaw) under the MIT licence.

Is DashClaw safe to use?

Our automatic scan flagged points to review, so read the code before installing it. Third party open source software. Sabemos AI does not maintain it. Check the code and permissions before you connect it to company data.