MCP dependency security checks for agent-kit

[Thezenmonster]

I ran a security scan on @inngest/agent-kit. Here's the report:
https://agentscores.xyz/report/%40inngest%2Fagent-kit

Score: 95/100. Clean posture, no critical findings.

Since agent-kit is an agent framework that works with MCP, the value here isn't flagging issues today. It's catching them before they reach users. When a dependency gets compromised or a new release introduces risk, the gate blocks the PR until someone reviews.

I built a GitHub Action for this. One YAML block:

- uses: Thezenmonster/mcp-verdict-action@v1
  with:
    api-key: ${{ secrets.AGENTSCORE_KEY }}
    fail-on: block

I'll provision a repo-scoped API key for inngest/agent-kit and configure everything. Free 30-day pilot. 5 minutes on your side.

Can I send you the key and workflow file?

Setup guide: https://agentscores.xyz/policy-gate
Action: https://github.com/Thezenmonster/mcp-verdict-action

[linear]

EXE-1609