Add agent security scan workflow

[luckyPipewrench]

Adds a GitHub Actions workflow that scans PRs for security risks using Pipelock.

Since this is a template repo, every project created from it inherits the workflow automatically.

What it checks:

• Exposed credentials, API keys, and tokens in diffs
• Prompt injection patterns
• Common security issues in agent codebases

How it works:

• Runs on every PR to main
• Read-only permissions (contents: read)
• Fails the check if it finds something
• No configuration needed

Single workflow file, no changes to existing code or CI.

[vercel]

@luckyPipewrench is attempting to deploy a commit to the Vercel Labs Team on Vercel.

A member of the Team first needs to authorize it.